mirror of
https://github.com/BerriAI/litellm.git
synced 2026-11-14 09:03:16 +00:00
1894 commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
Kerry Lu
|
07c7229895
|
feat(proxy): per-request header opts a stream into usage.cost without the global include_cost_in_streaming_usage flag (#50574)
* feat(proxy): let one streaming request opt into usage.cost with x-litellm-include-cost-in-streaming-usage The global include_cost_in_streaming_usage flag moved every stream off the fast path. A request can now send x-litellm-include-cost-in-streaming-usage: true to get usage.cost on its usage-carrying stream event (/v1/messages and the Anthropic and OpenAI pass-throughs) while requests without the header keep the fast path. The global flag keeps working unchanged. Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test: type the new stream cost header tests * fix(types): keep reportUnknownArgumentType at its merge-base count * fix(proxy): the stream cost header only accepts true --------- Co-authored-by: devin-ai-integration[bot] <158243242+devin-ai-integration[bot]@users.noreply.github.com> |
||
|
joshua-berri
|
3c3f266322
|
fix(router): a cancelled or retried attempt releases its io-token reservation and least-busy count (#50486)
* fix(router): a cancelled or retried attempt releases its io-token reservation and least-busy count A routed attempt cancelled mid-call (client disconnect with cancel_on_disconnect) kept its ITPM/OTPM reservation and its least-busy in-flight count, and retries sharing one logging object leaked the count of every attempt after the first. Each async attempt now runs under its own attempt id, so the attempt and its own callbacks release what it holds exactly once, in tasks a cancellation can't stop. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * fix(router): settle an io-token reservation once even through a logged copy of its metadata /v1/messages and /v1/responses stash the reservation in litellm_metadata, which the logging object copies. A cancelled attempt refunded it, then the proxy's failure logging refunded the copy again. Each stash now carries an id, and a reservation already settled under that id is not settled again. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * fix(router): keep spent tokens counted, refund a previous attempt's stash, and fit every call in flight A request cancelled after its provider answered now keeps the tokens it spent counted instead of refunding them. A retry refunds the reservation its previous attempt left behind even though it runs under a new attempt. The least-busy reservation store holds twice the keys, since each call in flight now keeps one more. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * fix(router): never evict a settled reservation id before its counters' window has passed Settled ids lived in a capacity-bounded cache, so past 10,000 settlements in two minutes a late copy of an evicted reservation was refunded a second time. They now leave only when their window expires. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * fix(router): start releases eagerly and keep attempt settlement out of router_utils A release that never waits (an in-memory counter) now finishes before its caller goes on, as it did when the caller awaited it inline, so a websocket turn's usage is counted by the time its logging is flushed. Each reservation carries its own id, and the attempt settlement module moves to litellm_core_utils, since litellm/utils.py may not import from router_utils. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * fix(router): count only finished answers as spent, and declare the least-busy redis family A stream's wrapper returning is not an answer: a stream cancelled before its first chunk now hands back its whole reservation. Only a non-streaming result marks its attempt as answered. The module level least-busy shared decrement carries its redis family, as every direct redis caller must. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> --------- Co-authored-by: Joshua Valluru <326636767+joshua-berri@users.noreply.github.com> Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com> |
||
|
yuneng-jiang
|
c024a4d2e3
|
test(proxy): delete 28 dead skipped tests in test_key_generate_prisma.py (#50550)
57 of the 62 tests in this file are hard-skipped because they need a live Prisma DB, so they never run. This removes the 20 whose behavior a running integration or e2e test already asserts, and the 8 that assert nothing meaningful or can never pass. The imports and the one helper only they used go with them. The other 29 skipped tests stay as a list of behavior still to port into tests/integration. |
||
|
Kerry Lu
|
432efc9b2d
|
fix(anthropic): exchange Google identity tokens again on refresh and read oidc/ references out of ANTHROPIC_IDENTITY_TOKEN (#50583)
A Google metadata-server identity token carries no jti, so Anthropic lets it be exchanged more than once. The shared on-disk token store still answered a mandatory refresh with the stored token minted from that same assertion, which was the almost expired token the refresh existed to replace. Google references now skip the store, so every refresh is a real exchange and a Kubernetes token file keeps today's cross-worker reuse. ANTHROPIC_IDENTITY_TOKEN was always wrapped as oidc/env/ANTHROPIC_IDENTITY_TOKEN, so a value that was itself an oidc/google/ reference was posted to Anthropic as the literal assertion text. A value starting with oidc/ is now the reference, with oidc/env_path/ still rejected, and a raw token in the variable resolves exactly as before. Fixes #50553 Co-authored-by: devin-ai-integration[bot] <158243242+devin-ai-integration[bot]@users.noreply.github.com> |
||
|
Kerry Lu
|
c5ad400f0b
|
fix(anthropic): refresh the WIF bearer token on the connection error retry (#50570)
* Refresh the WIF bearer token on the connection error retry * Only refresh the bearer when the retry goes to the host the token was minted for --------- Co-authored-by: devin-ai-integration[bot] <158243242+devin-ai-integration[bot]@users.noreply.github.com> |
||
|
Kerry Lu
|
51b49973c9
|
fix(proxy): hold DB_RECONCILE_LOCK around credential create, update and delete (#50551)
* fix(proxy): hold MODEL_RECONCILE_LOCK around credential create, update and delete The config reload snapshots the credential table, awaits config rows, then prunes and upserts litellm.credential_list from that snapshot. A credential write landing inside that window was undone in memory until the next reload. The three endpoints now take the same lock the reload holds. Fixes #50273 Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * fix(proxy): take MODEL_RECONCILE_LOCK before the credential update reads its row Two PATCHes queued behind a reload read the same old row and the second overwrote the first. The relay now keeps enough bytes to spot a split after marker, and the new unit tests are fully typed. * test(credentials): annotate the reconcile lock test locals with Final * refactor(proxy): rename MODEL_RECONCILE_LOCK to DB_RECONCILE_LOCK The lock gates every db-to-memory reconcile and now the credential writes too, so the model-only name undersold what it protects. --------- Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> |
||
|
yucheng-berri
|
b83a1f0608
|
fix(otel): name the routed model on auto-router LLM spans (#50498)
On an auto-router request the LLM span reported the name the caller asked for (the router or an alias of it) as gen_ai.request.model (OTel v2) and gen_ai.response.model (v1 and v2). Both now name the model group the router picked, read from the request's routing_decision. In v2 the span name and the vendor keys built from the request model (Langfuse, OpenInference, Langtrace, Weave) follow. Requests that were not auto-routed are unchanged. |
||
|
ryan-crabbe-berri
|
fa254b9346
|
fix(responses): surface mid-stream upstream errors that arrive without an event type (#50466)
* fix(responses): surface mid-stream upstream errors that arrive without an event type
Bedrock's native Responses route can end a stream with a bare {"error": {...}}
chunk. It parsed as a typeless GenericEvent and crashed the iterator with
AttributeError, so clients saw that instead of the upstream error. Such chunks
are now read as error events, which raise the mapped upstream error and reach
the client as response.failed.
* refactor(responses): inline the untyped error chunk check to keep LIT001 at the base count
|
||
|
yucheng-berri
|
b603e84c02
|
fix(mcp): derive MCP bridge and session keys with the FIPS provider's KBKDF (#50032)
* fix(mcp): derive MCP bridge and session keys with the FIPS provider's KBKDF With OpenSSL enforcing FIPS, hashlib.scrypt raises, so every MCP session token and bridge envelope request failed with a 500. Under FIPS the keys now come from LITELLM_SALT_KEY through SP 800-108r1 KBKDF run by the provider's EVP_KDF. Without FIPS the scrypt keys are unchanged * fix(mcp): keep master key rotation invalidating MCP tokens under FIPS Pass the master key as the KBKDF context, so rotating it changes the three FIPS-mode keys as it changes the scrypt keys * test(mcp): annotate new FIPS key test locals with Final * fix(mcp): derive the FIPS MCP keys only when general_settings.fips_mode is on On a FIPS OpenSSL 1.1.1 host (RHEL 8) scrypt still works, so keying the switch on openssl_enforces_fips() sent those hosts to the OpenSSL 3 only KBKDF and broke MCP sessions and bridge envelopes that worked on main. fips_mode_active() is the same reader B1 adds, set by enforce_fips_mode at startup. * refactor(fips): move the salt key KBKDF into encrypt_decrypt_utils The KBKDF helper derives keys from LITELLM_SALT_KEY, so it now sits next to the other salt key helpers, and fips.py goes back to main's version. * Pin the MCP key KBKDF fetch to the FIPS provider Fetch KBKDF with the property query fips=yes, so a change to the process's default properties after boot can't move the derivation to a non-FIPS provider. The shared fips_openssl test fixture drops the pin, since the test OpenSSL has no FIPS provider. |
||
|
ryan-crabbe-berri
|
02d610f1a9
|
fix(anthropic): disable optional default thinking for reasoning effort none (#50541) | ||
|
Yogeshwaran Ravichandran
|
8d9ddb349e
|
feat(guardrails): add prebuilt credential patterns for OpenAI, Anthropic, and Gemini (#50268)
* feat(guardrails): add prebuilt credential patterns for JWT, and OpenAI, Anthropic, Gemini API keys * fix(guardrails): prevent quadratic backtracking in jwt_token pattern * test(guardrails): move credential pattern tests into test_patterns.py * feat(guardrails): remove jwt_token to keep credential patterns strictly bounded * fix(guardrails): mask whole provider keys and keep openai_api_key to OpenAI's key shapes End each pattern so a key ending in "-" is masked in full, require OpenAI's T3BlbkFJ marker on bare sk- keys so other vendors' sk- keys and LiteLLM virtual keys are left alone, add sk-None- user keys, and replace the registration-only tests with whole-match, length-boundary, and negative cases * fix(guardrails): require the T3BlbkFJ marker on every OpenAI key form and match keys after escaped separators openai_api_key now needs the marker on the sk-proj-, sk-svcacct-, sk-admin- and sk-None- forms too, as gitleaks and TruffleHog do. anthropic_api_key and gemini_api_key also start a match right after a literal \n or \t or a percent-encoded byte, so keys in pasted JSON strings and URL-encoded logs are masked. * fix(guardrails): bound the OpenAI key regex, match Gemini prefixes case-sensitively, and hold back long keys in streamed replies * fix(guardrails): size the streaming holdback per pattern, stop OpenAI key tails before the next key, and allow non-ASCII text after a key * fix(guardrails): mask content filter pattern matches found in the original text so overlapping patterns cannot split a key * fix(guardrails): let a block pattern see the text after earlier masking patterns, as before Masking patterns still match the original text and merge overlapping matches into one tag, but a BLOCK pattern checks the text with the masking patterns listed before it applied, so a request main masked and passed is not blocked now. Detections name only the patterns whose tag was written. * fix(guardrails): keep content filter masks in order and hold back the longest OpenAI key Masking patterns run in configured order on the masked text as on main, so a contextual pattern still matches next to an earlier tag, and each also matches the original text so an earlier tag cannot split a key. Detections follow the configured pattern, not its name, and the OpenAI streaming holdback covers the pattern's 419-character longest match. --------- Co-authored-by: mateo-berri <277851410+mateo-berri@users.noreply.github.com> |
||
|
Kerry Lu
|
9ec3b5fd8b
|
feat(cost-map): move -latest model aliases into the cost map (#50491)
* feat(cost-map): move -latest model aliases into the cost map Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test: load the finalized local cost map in the moonshot alias test and assert alias pricing by behavior Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test: compare alias cost map rows to their targets Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * feat(cost-map): give each -latest alias its own alias_of row Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * fix(cost-map): keep alias_of rows out of alias lists and add them to the registry schema Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> (cherry picked from commit |
||
|
DDT
|
d0c16008c8
|
fix(chatgpt): round-trip encrypted reasoning through /v1/messages (#49076)
* fix(chatgpt): round-trip encrypted reasoning through /v1/messages The chat bridge stored encrypted reasoning on reasoning_items, the Anthropic adapter never read that field, stream chunks that only carried thinking blocks were dropped, and the messages adapter stripped the signed blocks before a chatgpt request could turn them back into reasoning items. Co-authored-by: DDT <1786035110@qq.com> * fix(anthropic): keep Responses reasoning on tool turns and retry once when a backend cannot decrypt it * fix(anthropic): stop logging rejected encrypted reasoning Quote the retry helper return type so importing litellm no longer evaluates a string alias with |. The warning keeps provider, model, status, and error code, and leaves the ciphertext out. * fix(anthropic): keep encrypted reasoning on OpenAI thinking and xAI web search turns, and stream signatures only to Messages clients The keep check now knows when the handler moves an OpenAI thinking turn to Responses, and when web_search_options sends an xAI turn there. The Responses-to-chat stream iterator adds the signed reasoning chunk only when the caller is /v1/messages, so /v1/chat/completions streams stay unchanged. The reasoning retry fires only when the translated request still carries encrypted reasoning. * fix(anthropic): keep each streamed reasoning summary with its own item, and read the Messages marker before the bridge call * fix(anthropic): send openai/ Messages turns again once without encrypted reasoning a backend rejects The openai/ Responses path on /v1/messages had no retry, so a chatgpt/ block left in history answered 400 on every later openai/ turn. That path now sends the turn once more without encrypted reasoning, the same way the chat bridge does. The retry matches only invalid_encrypted_content or a "could not be verified" message naming the content, so the exception mapper's own hint no longer triggers it. The xAI web search keep path is backed out, since xAI sends no encrypted reasoning over the bridge. * fix(anthropic): keep every non-streamed reasoning item for Messages clients and read a deployment's reasoning_summary in the keep check * fix(anthropic): log only the outcome of a turn sent again without rejected reasoning A /v1/messages turn whose replayed reasoning the backend rejects is sent once more without it. The rejected first attempt ran the failure callbacks on the shared logging object, which logged a failure for a turn that then succeeded, kept the retry's own failure from being logged, and freed the parallel-request slot while the retry was still in flight. The first attempt now runs inside resent_on, and the client wrapper skips failure logging and the budget unbind for a failure that matches it. The retry runs outside that context, so its own outcome logs as usual. * test(anthropic): type the parameters of the failure-recording test loggers * fix(anthropic): skip the SDK num_retries retry for an attempt that is sent again * fix(anthropic): stop SDK retries from resending a rejected block or clearing the global num_retries A transient 429 or 500 on a /v1/messages turn that carries a replayed encrypted block was retried by an SDK call's num_retries loop, which sent the block again until every retry was rejected. On async calls the loop then raised the first transient error, so the turn failed instead of being sent again without the block. The retry loop now stops at that rejection and passes it on, so the turn is sent again once without encrypted blocks. A rejected first try on a sync call also cleared the process-wide litellm.num_retries, so later calls in the process were no longer retried. That setting now stays as configured. * fix(anthropic): stop a cancelled SDK retry from starting another call The resent-failure retry predicate returned true for BaseException subclasses such as asyncio.CancelledError, so a cancelled acompletion_with_retries or aresponses_with_retries call could start another provider call. Retry only Exception subclasses again, as tenacity's default did, while still skipping a resent failure. * test(anthropic): type the retry helper parameter in the cancellation test --------- Co-authored-by: mateo-berri <277851410+mateo-berri@users.noreply.github.com> |
||
|
Kerry Lu
|
ffc8d6cf66
|
feat(cost-map): read alias_of pointer rows in the catalog and the loader (#50512)
* feat(cost-map): read alias_of pointer rows in the catalog and the loader Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * fix(cost-map): keep alias_of rows out of alias lists and add them to the registry schema Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * fix(cost-map): let alias lists win over alias_of rows in both loaders and never serve a malformed pointer Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> --------- Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> |
||
|
ryan-crabbe-berri
|
3a170ca1fb
|
fix(utils): drop Anthropic-only output_config under drop_params on providers that do not read it (#50542) | ||
|
Mateo Wang
|
1f98627d4a
|
refactor: declare each lazy litellm attribute once and resolve them through one function (#50523) | ||
|
Mukesh Dewangan
|
4ed339b2fc
|
fix(vertex_ai): map missing image on Gemini image edit to BadRequestError (400) (#50244) | ||
|
Mateo Wang
|
709b27e268
|
fix(anthropic): preserve streamed tool argument fragments verbatim (#50470)
* fix(anthropic): preserve streamed tool argument fragments verbatim Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(anthropic): send the schema-less stream test the null properties form providers accept * fix(anthropic): unwrap values when a schema-less stream sends its whole tool input in one piece * fix(anthropic): unwrap only a one-piece values input and never emit Infinity in schema-less streams * fix(anthropic): keep JSON mode answers verbatim when they cannot be written back as strict JSON A tool input holding a number past a float or an integer past Python's digit limit now passes through as the model wrote it on the stream and non-stream paths, instead of coming back with Infinity or raising. One-piece values rows for false, null, a plain object and too-deep nesting pin the remaining unwrap branches. * fix(anthropic): reject non-finite floats through pydantic instead of a raising parse hook * fix(anthropic): keep split JSON mode answers whole when a field cannot be written back as strict JSON * fix(anthropic): keep out-of-range numbers as the model wrote them in JSON mode answers Decode tool input with parse hooks that keep a number past a float or an integer past the digit limit as its source text, and write it back unchanged, so held, wrapped and name and value split answers stay whole instead of falling back to raw blocks. * fix(anthropic): keep numbers that round to zero as the model wrote them in JSON mode answers * refactor(anthropic): encode JSON mode answers without a recursive helper * fix(anthropic): decode deeply nested JSON mode answers without slow recursive validation * fix(anthropic): type the decoded JSON mode answer as object and test non-object answers json.loads output now goes through the existing TypeAdapter(object), which returns it unchanged, so no Any reaches the object guard. New stream and non-stream cases check that a list, string or number answer comes back as written. --------- Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> |
||
|
yujonglee
|
574fcea691
|
fix(redis): share one native circuit breaker between RedisCache and the Rust stores (#50473)
* fix(redis): share one native circuit breaker between RedisCache and the Rust stores RedisCache and the native Redis client each kept their own breaker, so once the cache learned the coordination Redis was wedged, a spend-counter write through the native store still paid a full socket timeout. Every Redis target now has one Rust breaker, looked up by NativeRedisBreaker.shared and injected into both the RedisCache guard and the native client. The Python FSM is gone. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * style(rust): cargo fmt the shared breaker Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * fix(redis): close the breaker when the recovery probe gets a Redis reply error A command error such as WRONGTYPE means Redis answered, but the probe was left unsettled and its drop reopened the breaker for another recovery timeout. Both the RedisCache guard and the native client now settle such a probe as a success. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * test(redis): pass a NativeRedisBreaker and isolate the shared breaker per test The sensitive routing test still passed RedisBreakerSettings as the store breaker, and the session affinity fixture reused one Redis URL, so a breaker an earlier test opened leaked into the next. Also make the RUF043 match pattern raw. * fix(redis): key the shared breaker by server and close the three audit gaps The breaker registry now keys on the Redis server (host and port, unix socket, cluster node set or sentinel service) instead of the full connection target, so caches that differ only in db, credentials or url spelling share one breaker. A READONLY reply from a replica mid-failover no longer counts as Redis answering, so it can't reset a run of failures. A new RedisCache refreshes the breaker state gauge, which the deleted Python breaker did at construction. A shared native client is only reused by a store that passes the same breaker. * refactor(redis): split redis-types into modules and drop redundant pyo3 attribute annotations lib.rs now only declares modules and re-exports, with resolution, target, tls, auth and endpoint each in their own file. Derived FromPyObject already reads a struct field as the attribute of the same name, so #[pyo3(attribute)] on the breaker and store settings did nothing. * refactor(redis): drop redundant pyo3 attribute annotations in the stores and let PyO3 build the cooldown snapshot Derived FromPyObject already reads each field as the attribute of the same name. The cooldown snapshot returns an IndexMap, which PyO3 turns into the same dict in the same order. * refactor(redis): return a HashMap from the three memory-tier snapshots The snapshots read cache-memory's HashMap, so there is no order to keep. PyO3 builds the same dict from a std HashMap, which replaces the hand-written PyDict loops in the spend counter and router usage stores and the IndexMap in the cooldown store. The rate-limit slot registry keeps IndexMap because it hands back a Python dict in the order Python wrote it. * refactor(redis): share Redis breakers and clients through one RedisHosts per host redis-connect's RedisHosts now owns the per-server breaker and the shared clients that python-bridge kept in two process-wide statics. A client takes breaker settings and admits through its server's breaker, so no caller can pair a client with a breaker of its own, and clients are keyed by Resolution alone. This drops RedisClient::guarded_by and Breaker::same. python-bridge keeps one RedisHosts per process and calls it with the GIL released. The gateway builds one per connect, so its coordination and response cache on the same server now share a breaker. RedisStoreSettings.breaker is back to breaker settings, as on main. --------- Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com> |
||
|
quietjules
|
210e2a075a
|
fix(vertex_ai): authenticate Vertex AI pass-through requests once (#48942)
* test(vertex_ai): run auth once per Vertex pass-through request Drive /vertex_ai, /vertex-ai and /vertex_ai/discovery requests through the real proxy app with a counting custom_auth, and assert each request authenticates once and is served as the caller that auth returned. The generate-content routes fail today because the route authenticates a second time. * fix(vertex_ai): authenticate Vertex pass-through requests once vertex_proxy_route already resolves the caller through Depends(user_api_key_auth), but _base_vertex_proxy_route ignored it and authenticated the request again. A custom_auth function ran twice per request, and a key with max_budget reserved its estimated cost twice, so the unreleased first reservation exhausted the budget after one or two calls. The base route now uses the caller it is given and authenticates once itself only when none is passed, which is the discovery route's case. * fix(vertex): read the caller's key only from x-litellm-api-key or Authorization on Vertex routes The single auth check on /vertex_ai and /vertex-ai now reads the same two headers main's effective check did, so a key sent only in another key header gets 401 again instead of authenticating. * fix(vertex): authenticate the discovery route through the same Vertex key check Both Vertex pass-through routes now resolve the caller once through _authenticate_vertex_caller and hand it to _base_vertex_proxy_route, which no longer re-runs auth. x-litellm-api-key accepts the same scheme prefixes on /vertex_ai/discovery as on /vertex_ai and every other proxy route. * fix(proxy): size a Gemini request's budget hold from its prompt and maxOutputTokens A Gemini generateContent body's budget hold now counts its contents and systemInstruction and reads generationConfig.maxOutputTokens. Before, it held the model's whole context window, about $0.85 on gemini-3.8-flash, so a key with a small budget got 422s on back-to-back calls it could afford Also removes get_litellm_virtual_key, which nothing calls since the Vertex routes stopped using it * fix(proxy): count every Gemini system part and candidateCount in the budget hold A Gemini generateContent body with several systemInstruction text parts lost every part after the first when LiteLLM translated it to a chat request. That dropped the later system text for non-Gemini models behind /v1beta/models/<model>:generateContent, and made the budget hold count only the first part. The adapter now keeps every text part, as a plain string when there is one part and as text blocks when there are several. The output hold now also multiplies by generationConfig.candidateCount, the same way chat bodies use n, so a request asking for several candidates holds budget for all of them. * fix(proxy): keep the context-window hold for Gemini bodies the prompt count cannot see A Gemini generateContent body with media, cachedContent, other roles, or non-function tools keeps the full context-window hold, since the chat translation drops or undercounts those parts. The hold also reads the snake_case generation_config output fields, and several systemInstruction text parts reach non-Gemini models joined into one system string * fix(proxy): read only camelCase Gemini output fields in the budget hold /v1beta forwards only generationConfig.maxOutputTokens to non-Gemini models, so a snake_case cap the model never receives must not shrink the hold. Snake_case bodies now get the default output hold. * fix(proxy): count snake_case candidate_count in the Gemini budget hold * fix(vertex_ai): skip malformed system parts and read Gemini output fields only on Gemini bodies A systemInstruction part whose text is not a string is skipped, as main did, instead of failing the request. The budget hold reads generationConfig output and candidate fields, and counts contents, only when the body has no messages, prompt, input, query or documents field. An output cap of 0 or below counts as no cap. Test recorders keep tuples instead of mutating lists. * fix(proxy): take the budget hold's request shape from the route On generateContent routes the hold counts only contents and reads its output cap and multiplier from generationConfig, so stray fields of other APIs can no longer hide the prompt. A max_tokens of 0 holds no output again, while negative values and a Gemini maxOutputTokens of 0 or below hold the default output * fix(proxy): hold one candidate for a count too large to price and read system parts like content parts * fix(proxy): keep the spend already counted when a budget hold is far over the budget * fix(proxy): count candidateCount in the budget hold only for models that generate the candidates A non-Gemini model reached through /v1beta generateContent never receives candidateCount, so a huge count held the whole budget for the length of the call while the model produced one answer * fix(proxy): hold a candidate count only when every deployment generates it, and drop n when every deployment drops it * fix(proxy): count candidateCount when any deployment generates it and always count n The stricter rules from the previous commit let a key spend past max_budget through a Gemini name no deployment serves, an allowed_openai_params n, and a router fallback. Restore the earlier rules and add regression tests for those three shapes. * fix(proxy): cap the candidateCount hold at main's default output hold A candidateCount like 1e300 sent to a mixed Gemini and Claude group, or through a key alias from a Gemini name to Claude, held a whole key while Claude served the call and answered a strict key 422. On generateContent routes the counted candidates now raise the output hold at most to the default output hold main used for the model, so no count makes a call hold more than on main. Chat n is unchanged. * fix(proxy): hold every Gemini candidate up to the merge base's whole hold The previous cap stopped the candidateCount output hold at the default 16384 tokens. Because the hold now counts the Gemini prompt instead of the whole input window, a strict key admitted a call like 4 candidates of 8000 tokens and then spent past max_budget, where the merge base refused it. Every candidate is now held at its output cap, like chat n, and the whole hold is clamped at what the merge base held for the same call (the model's input window plus the default output), never below one candidate's hold. A huge count still holds no more than the merge base did. Chat n is unchanged * fix(proxy): let a Gemini output cap that /v1beta drops only raise the hold The Vertex and Gemini pass-throughs forward every spelling of the output cap, but /v1beta passes only generationConfig.maxOutputTokens to other models. The hold read only that spelling, so a snake_case cap above the default output held less than the pass-through could spend. A cap in any other spelling now raises the hold to it and never lowers it * fix(proxy): hold Gemini candidates of a model with no context window up to the default output A Gemini model the cost map has no context window for, priced in its deployment, held one candidate for any candidateCount, so parallel calls could pass their budget holds. Its candidates are held again up to its counted prompt plus the default output hold * fix(proxy): keep main's Gemini hold unless the request caps its output An uncapped generateContent call, or one capped only by a snake_case spelling, now holds what main holds (the context window plus the default output). A call with generationConfig.maxOutputTokens holds its counted prompt plus cap times candidateCount, never clamped to the deployment's max_output_tokens and never more than main's hold. A model with no context window keeps main's no-hold, and candidateCount always counts on generateContent routes. Also keeps get_litellm_virtual_key in pass_through_endpoints/common_utils.py, so an out-of-tree importer does not break * fix(proxy): keep main's budget hold on generateContent routes The PR now leaves the generateContent budget estimate as main has it. A live risk check showed the counted Gemini hold under-reserves (the prompt counter undercounts Gemini bodies, tiered and image output rates are not read, and request fields the handler merges later are not counted), so that estimate moves to a follow-up. The single auth pass on the Vertex pass-through and the systemInstruction adapter fix stay * fix(vertex_ai): keep main's Gemini adapter and pass a missing Authorization as no key The Gemini adapter change for later systemInstruction parts goes back to main, since it is not the double-auth fix this PR is for. The Vertex auth dependency now hands a missing Authorization header to the auth check as no key, like main's first auth pass, so custom auth and the 401 text see what they see on main --------- Co-authored-by: mateo-berri <277851410+mateo-berri@users.noreply.github.com> |
||
|
devin-ai-integration[bot]
|
0eb6816081
|
fix(gemini): add Gemini 2.5 Pro flex cache price above 200k tokens (#50480)
* fix(gemini): add Gemini 2.5 Pro flex cache price above 200k tokens * test(cost-map): allow cache_read_input_token_cost_above_200k_tokens_flex in the row schema --------- Co-authored-by: devin-ai-integration[bot] <158243242+devin-ai-integration[bot]@users.noreply.github.com> |
||
|
Ankit Jha
|
72228db5c6
|
fix(caching): don't cache or replay a reply that has no output (#50057)
* fix(caching): don't cache an Anthropic Messages stream whose content blocks are all empty
A non-Anthropic model routed through the Messages adapter turns an empty reply into a single empty text block. The no-output check counted any content_block_start as output, so the async stream cache stored that empty reply and replayed it on every repeat. Judge each content_block_start and content_block_delta by its payload instead, like chat streams judge a choice: a block carries output when a field other than type holds something besides "", [] or null. Empty replies are no longer stored and stored ones become misses. A bare {} is not treated as empty, since the only block fields that hold one (tool_use input) come with an id and a name
Fixes #49931
* fix(caching): treat thought-signature-only and non-streamed empty replies as no output
* fix(caching): treat whitespace-only replies as no output
* fix(caching): keep whitespace replies and text completion fills cached
* refactor(caching): drop the unused streamed-choice branch of the output check
---------
Co-authored-by: mateo-berri <277851410+mateo-berri@users.noreply.github.com>
|
||
|
HootRock
|
8f19607259
|
fix(proxy)!: preserve Responses SSE event framing (#47980)
* fix(proxy): preserve Responses SSE event framing
* style(proxy): format Responses stream terminator condition
* fix(proxy): validate Responses guardrail event names
* fix(proxy): keep Responses first-event errors and keepalive failures in Responses framing
* test(proxy): use controlled clocks for Responses keepalive failures
* fix(proxy): end late Responses failures with an error event the SDK raises on
Errors that arrive after keepalive pings, and guardrail error strings yielded
mid-stream, now end /v1/responses with a named `error` event carrying the
same error object chat sends, so the OpenAI SDK still raises APIError with a
clean message and provider_specific_fields. Background polling passes the
same framing and records an `error` event as a failed response. The first
data line of a frame is found by splitting on newlines only, so a U+2028 in
an error message no longer hides a first-event error.
* fix(proxy): match OpenAI's Responses error event and stream late passthrough blocks as block responses
* fix(proxy): keep the reason when a background Responses request fails
A background poll of a request a post-call guardrail blocked saved a
null code and message, because the guardrail's error event nests them.
A request whose first chunk was an error polled as completed with no
error. Both now poll failed with the same code and message a late
failure gets.
* test(proxy): type the first-chunk poll test parameters
* fix(proxy): keep first-chunk error detection to lone error frames and handle string poll errors
* fix(proxy): keep the HTTP status on late Responses errors and end Responses streams on error events
A late error body without a code now takes the response's status, an upstream error event counts as the stream's terminal event, and a Responses stream that fails before its first chunk ends with an error event instead of data: [DONE].
* test(proxy): type the first-chunk failure test's parameters
* fix(proxy): keep cursor chat framing and JSON-encode Responses hook errors
* fix(proxy): encode non-dict Mapping Responses chunks as JSON objects
* fix(proxy): encode mappings and models nested in Responses stream chunks as JSON objects
* test(proxy): freeze the hook-output fixture
* fix(proxy): keep masked mappings and odd keys safe in Responses stream frames
Only read-only dict views and pydantic models nested in a hook chunk or error event expand into JSON objects. Other mappings, such as httpx.Headers, keep their own string form so masked values stay masked. Keys JSON cannot take become strings, and a chunk JSON still cannot encode goes out as the plain data frame it was before.
* fix(proxy): never fail a Responses stream frame on a value JSON cannot encode
Only dicts and read-only dict views expand into JSON, at the top level too, so a header mapping yielded as a whole chunk keeps its masked string form. Any failure while encoding a chunk falls back to the plain data frame the base sent, and an error event that cannot encode still goes out with its code and message, so a late block never drops the connection.
* fix(proxy): keep what a nested value's repr hides out of Responses stream frames
* fix(proxy): print values in Responses stream frames the way each base path did
* fix(proxy): send string and number subclasses whose repr shows their value as that value in Responses stream frames
* fix(proxy): print container subclasses and SDK models in Responses stream frames the way base did
* fix(proxy): read hook container storage the way repr does and send error-event models as str()
* fix(proxy): frame streamed Responses reads and send unreadable hook values as base did
Track pre-framed and raw SSE frames so a late failure never follows a terminal event, accept CRLF and CR line breaks in a first named error, frame GET /v1/responses/{id}?stream=true as Responses events, print a whole hook proxy chunk once and fall back to its text, ignore a non-integer status_code in a hook error frame, read a late-block dict subclass through items() when that view differs from its storage, and keep the order and inline form base's repr printed for nested mappings
* fix(proxy): match base on Responses stream edge cases
Read a decimal-string or integral-float status_code as its integer again, so
"429" and 429.0 still map to rate_limit_exceeded. Print each hook value's repr
once and reuse it, so a repr that changes between calls cannot reach the
client. Track only pre-framed sequence numbers from 0 to 2**63-1, so a huge
one cannot break the failure event. Build nested JSON with one frame per level
again, which restores the nesting depth sent as JSON. Count a pre-framed frame
as terminal from its data type, and a named error frame only when its data
carries an error, the way the OpenAI SDKs read them.
* fix(proxy): print hook proxies once and keep base's chunk tracking
Read-only mappings in hook chunks now emit the text from their first print, hook
chunks are read without their error field, only a non-empty error object makes a
framed error event terminal, and framed sequence numbers stay below 2**63-1 while
hook chunks keep base's unbounded tracking
* fix(proxy): print each read-only mapping item once and poll framed error events as failed
A MappingProxyType in a hook chunk or error detail now expands from the
mapping it wraps, so each item is printed once, the way base printed it.
A value whose repr changes between calls stays masked, a value whose repr
works once is kept, and deep nesting keeps one stack frame per level
The background reader takes each frame's type from the same terminal rule
the stream uses, so an event: error frame with an error object polls
failed even when its data names another type or none
* fix(proxy): print self-holding hook values once and poll how the stream ended
A container that holds itself now goes out as the marker its repr prints, like {...}, [...] or mappingproxy({...}), and the dict behind a read-only mapping joins the cycle check, so no item is printed a second time. A dict whose item repr adds a key keeps going after the items already shown, the way dict's repr does, instead of printing the whole dict again
A whole read-only mapping chunk that cannot go out as JSON is printed once by the formatter, so a print that fails fails the stream as it did on the base
The background poll reads the event type from the block it parses, as on the base, and keeps the first error frame the stream treats as terminal anywhere in a chunk. That frame fails the poll only when no response.completed, response.failed or response.incomplete follows it
* fix(proxy): name hook chunk events from the sent JSON and keep polls on frames they cannot read
A mapping chunk now takes its event: line from the JSON it sends, so a masked type value never reaches that line, and a chunk the event model cannot read goes out as a plain data frame instead of failing the stream
A dict whose item repr deletes or swaps keys keeps printing every key not yet shown, and an error code past the integer digit limit reads as no status
The background poll skips a frame whose shape it cannot read, as base skipped frames it could not parse, stops reading a chunk at [DONE], and lets only the last terminal frame in a chunk decide its status
A Responses stream's first chunk answers as a JSON error only for codes 400 and up
* fix(proxy): read every frame of a polled chunk and name model chunks from the sent JSON
A background poll parsed only the first data frame of each chunk, so a terminal event behind an error or delta frame in the same chunk was lost, and the poll stored a status its own stream never ended with. It now reads every frame up to [DONE]
A terminal event whose response is a string or null kept no status and polled completed. Its status now comes from the event type, and final output items that are not objects are skipped, so the rest of the final output is still stored
Model chunks now take their event line from the JSON they send, as mapping chunks already do, so a serializer that drops type no longer sends event: response.completed over {}
* fix(proxy): read polled chunks as SSE events and keep draining past a mid-chunk [DONE]
A background poll split a chunk into data lines, so one event with two data lines parsed as two payloads, and a frame with an integer past the digit limit raised a ValueError the parse guard did not catch, which failed the poll and dropped its spend row. The poll now joins each event's data lines the way an SSE reader does and skips any frame it cannot parse
A [DONE] behind another frame in one chunk ended the poll and left the stream unbilled. Only a chunk that opens with [DONE] ends the poll now, as on main, and the poll stops reading a chunk at its first terminal event, so two terminal events in one chunk poll the first one as main did
* fix(proxy): JSON-encode only plain hook chunks on Responses streams
A hook chunk whose whole tree is builtin JSON types streams as its named
event. Any other chunk keeps the base data frame, so nothing a custom
repr hides reaches the wire. Error event details encode the way the chat
error frame does.
* fix(proxy): skip containers already visited when checking hook chunks for JSON
* fix(responses): keep streaming past an event with no type
The Responses stream iterator read event.type directly when it noted a
yielded event, so an event whose JSON had no type, like a custom hook's
dict that the proxy now sends as JSON, raised AttributeError and ended
the stream. Read it with getattr, as the iterator's other checks do.
* fix(proxy): keep hook dicts with no type on the base Responses frame
A hook dict with no type is not a Responses event, so it gets no event
name either way. Sending it as JSON only made litellm SDKs that read
type on every event stop the stream. Keep such dicts on the frame the
base sends, so every SDK version reads them as before.
* fix(responses): keep a background poll's first terminal status and take a first-chunk error code from its HTTP status
---------
Co-authored-by: mateo-berri <277851410+mateo-berri@users.noreply.github.com>
|
||
|
Krrish Dholakia
|
b3c4a1c3e4
|
feat(spend-logs): turn off prompt storage per key or team (#50135)
* Add per-key and team prompt storage restrictions for spend logs Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * Apply spend-log prompt restrictions to passthrough requests Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * Apply team prompt storage restrictions to queued requests Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * fix(spend-logs): honor tenant prompt storage in the collector and on "false" strings The spend-event collector decided whether to ship request and response bodies from the global setting alone, so an opted-out key or team still sent prompts through the collector transport. It now uses the same key/team check as the spend log writer. A key or team setting of the string "false" was ignored and prompts kept being stored. It now counts as false, the same way the global setting parses strings. Splits the 42-case tenant storage test into one test for the policy and one for the pass-through logging contexts. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * fix(spend-logs): carry key/team prompt storage into failure, batch and side-call rows Failure rows, executed batch rows, semantic index embeddings, websearch interception calls and errors rejected before request setup built their logging metadata without the key and team metadata, so a restricted key or team still had prompts stored on those rows. They now stamp it through one helper, LiteLLMProxyRequestSetup.get_tenant_logging_metadata. The dashboard toggle now shows a saved string "false" as off, matching the server, and its label and help text come from the translation file. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> --------- Co-authored-by: Krrish Dholakia <krrishdholakia@berri.ai> Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> Co-authored-by: Joshua Valluru <326636767+joshua-berri@users.noreply.github.com> Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com> |
||
|
Nate Armstrong
|
a3bf1f4399
|
fix(router): record retries and fallback hops past a request's first logged failure (#50445) | ||
|
devin-ai-integration[bot]
|
e9b1c11098
|
fix(xai): honor deployment image pixel pricing (#50490)
Co-authored-by: devin-ai-integration[bot] <158243242+devin-ai-integration[bot]@users.noreply.github.com> |
||
|
devin-ai-integration[bot]
|
94861ecdf7
|
feat(xai): bill image calls from cost_in_usd_ticks and add grok-imagine-image-2.0 resolution prices (#50481)
* feat(xai): use billed image cost and resolution prices * test(xai): mark image generation tests as a package --------- Co-authored-by: devin-ai-integration[bot] <158243242+devin-ai-integration[bot]@users.noreply.github.com> |
||
|
ryan-crabbe-berri
|
6b745635c8
|
fix(bedrock): forward Codex agent messages that carry encrypted content intact (#50457)
* fix(bedrock): forward Codex agent messages that carry encrypted content intact Codex multi-agent v2 sends worker messages as agent_message items whose encrypted_content block only the model server can decrypt. Bedrock Mantle and Bedrock Runtime used to 400 on agent_message, so LiteLLM rewrote every one into assistant text, pasting the ciphertext into the prompt and dropping the sender and recipient. Bedrock now decrypts these items itself, so the rewrite left the recipient model reading an opaque token instead of its task. Agent messages with an encrypted_content block now go to Bedrock untouched on both endpoints. The Bedrock configs declare supports_encrypted_agent_messages, and the shared normalizer keys off that flag so the two cannot drift. Mantle's /v1/responses route for open-weight models still rejects agent_message in any form, so it keeps the old rewrite. Plaintext-only agent messages keep the rewrite on every route. * fix(bedrock): only keep worker messages locked for OpenAI GPT models Gemma and Grok on Mantle's OpenAI route never lock Codex worker messages, so Codex puts plain text in the locked slot and Bedrock rejected it with invalid_encrypted_content once the route alone decided to forward it intact. Decide from the model instead, and restore the full-value rewrite assertion. |
||
|
Mateo Wang
|
98acc01e0d
|
feat(videos): add first-class OpenRouter video request fields (#50319)
* feat(videos): add first-class OpenRouter video request fields Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * fix(videos): remove recursive image parameter fallback Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * fix(videos): preserve EdenAI native parameter passthrough Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * fix(videos): deduplicate frame roles in linear time Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * fix(videos): preserve image selection and dropped parameter semantics Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * fix(videos): preserve provider scalar parameters and raw extra body Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * fix(videos): retain accepted scalar inputs for Vertex and DashScope Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> --------- Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> |
||
|
mrinal-berri
|
e3387b6fc5
|
fix(otel): seed request metadata baggage for team-only OTel v2 destinations (#50373)
* fix(otel): seed request metadata baggage for team-only OTel v2 destinations A team newrelic (or other preset) destination exported spans without the operator-allowlisted litellm.metadata.* keys unless a proxy-wide otel callback was listed, because only the OTel logger's async_pre_call_hook seeded requester_metadata into Baggage and that hook runs only for loggers in litellm.callbacks. ProxyLogging.pre_call_hook now re-seeds identity with the request data once requester_metadata exists. Resolves LIT-9783 * fix(otel): seed only request metadata from the proxy pre-call hook The pre-call re-seed also passed the request model, which by then can be an alias target, mcp-tool-call or an internal embedding model, so it rewrote gen_ai.request.model on the server span for V2 deployments without otel listed. Seed only the request metadata and keep the model auth seeded. Adds a unit test for the pre-call seed and the guardrails_only skip, drops the explanatory comments, and requires one server span plus the model span in the integration test. * test(otel): annotate new test variables Final and build test config without mutation Addresses review: the new unit and integration tests now follow AGENTS.md (LIT010 Final on every variable, no in-place mutation, read-only parameter types, one for-clause per comprehension). * test(otel): cover key-level callbacks and assert tenant spans carry no credentials Parametrizes the tenant metadata test over team and key callbacks, and adds a test that sends credential headers and a planted New Relic key with an allowlist naming credential headers, then asserts no exported span carries the master key, the virtual key, the team's New Relic key or any credential header value, and that every export used only the team's own New Relic key. |
||
|
yujonglee
|
2b4485dee5
|
fix(guardrails): check a model-level post_call response once across the proxy and deployment layers (#50300)
#40154 made the deployment post_call hook really run apply_guardrail guardrails, so the proxy checked the same response a second time (LIT-10481). The deployment hook now records guardrail + response hash on the request and on the response, and the proxy skips a guardrail only on a matching record. The hash covers every field except ids, usage, model and other metadata set between layers, so a response rewritten between the layers (OCR included) is still checked. Guardrails with their own post_call hook keep their pre-#40154 behavior Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com> |
||
|
devin-ai-integration[bot]
|
5d66f2d8ff
|
fix(usage): keep spend when metadata lookups hit database failures (#50253)
* fix(usage): tolerate entity metadata database failures Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * fix(usage): keep spend when metadata lookups hit database failures Treat transport errors and the query engine crash on a closed connection as database failures in the key, user and guardrail lookups of the aggregated, keys, users, model_top_keys and cache_leakage_keys routes, and keep prisma out of module-level imports so a proxy install without prisma still imports the usage endpoints. * fix(usage): fall back only on database outages, on every daily activity route Data-layer Prisma errors such as a rejected query or a missing table return 500 again, a key whose details failed to load reports key_exists null instead of false, and the non-aggregated daily activity routes keep spend when the key or entity metadata lookup hits an outage * fix(proxy): skip the key metadata lookup when the daily rows carry no keys * fix(usage): keep the guardrail breakdown fallback for any Prisma error A missing guardrail usage table or column made the Usage call return 500 after the outage-only narrowing. The guardrail lookup now falls back on outages and on any Prisma error, as it did before this PR * fix(usage): keep spend when the team, organization, customer, or agent name lookup drops --------- Co-authored-by: mateo <277851410+mateo-berri@users.noreply.github.com> Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> |
||
|
ryan-crabbe-berri
|
ab3b12b951
|
feat(proxy): serve Codex its own release's model catalog, fetched from GitHub (#50318)
* feat(proxy): give Codex accurate model entries for every model in /v1/models * fix(router): read a deployment's upstream model without a not-required key access * fix(proxy): keep Codex tiers for unpriced deployments and match stock entries by declared provider A deployment with no token price no longer votes a Codex service tier off, an undescribed model keeps Codex's default reasoning summary parameter, a Bedrock deployment naming its provider in custom_llm_provider gets Codex's stock entry, and Bedrock provider namespaces follow the AWS model id format. * fix(proxy): import cost tracking and provider helpers lazily in the Codex catalog lite codex imports the catalog before the SDK is initialized, so importing cost tracking, the Bedrock helpers or provider resolution at module level hit a circular import and broke every lite command. * fix(proxy): drop Codex's fast speed tier wherever its priority tier is dropped Codex shows /fast for a leftover fast speed tier even when service_tiers no longer offers priority, so an alias, an empty configured tier list or a configured list without priority showed a /fast toggle that sends nothing. One helper now sets service tiers, speed tiers and the default together. * fix(proxy): give an older Codex a translated entry for models past its version Codex's own entries name the oldest release that can read them in minimal_client_version. A release older than that can fail to decode the entry, such as a reasoning effort it has no name for, which fails the whole catalog, so it now gets the translated entry, whose levels are trimmed to that release * feat(proxy): serve Codex its own release's model catalog, fetched from GitHub When Codex asks /v1/models for its catalog, the proxy now reads models.json and prompt.md at that release's tag on GitHub and serves those entries, falling back to the bundled 0.166.0 copy when the fetch fails, times out after 2 seconds, does not validate, the version has no tag, or LITELLM_OFFLINE_MODE is on. A failed fetch is remembered per version for an hour and only the eight most recent versions are kept. The integration tests ask as the bundled release so their expectations hold whether or not the proxy can reach GitHub * fix(proxy): bound the GitHub fetches a caller can trigger through Codex catalog versions Failed versions now live in their own cache so they never push out a fetched catalog, requests for a version already being fetched share that fetch, each worker starts at most six fetches a minute and serves the bundled catalog past that, and the version check only accepts ASCII digits * fix(proxy): keep a failed Codex version's cooldown past the fetched catalog cap Failures are tiny, so they get their own cap of 256 instead of sharing the eight-catalog cap, which let nine failing versions retry each other well before their hour was up. The TypeVar takes Final like the rest of the codebase |
||
|
Mateo Wang
|
3f38730d2c
|
refactor(typing): clear about 1,100 basedpyright Any errors across 303 files (#49986)
* refactor(typing): clear basedpyright Any errors (wave 1) * refactor(typing): clear basedpyright Any errors (wave 2) * refactor(typing): clear basedpyright Any errors (wave 3) * refactor(typing): clear basedpyright Any errors (wave 4) * refactor(typing): clear basedpyright Any errors (wave 5) * refactor(typing): clear basedpyright Any errors (wave 6) * fix(typing): keep the cached-plan fallback's direct query_first calls so span naming resolves * fix(logging): keep callbacks running when litellm_params is a params model should_run_callback and the sync success and failure handlers read litellm_params through any object with a get method again, as they did before this branch, so a logging hook that returns a GenericLiteLLMParams no longer stops sync logging for that request Adds behavior tests for the changed lines codecov listed as not run * fix: read the regional uplift for every data residency again and reuse one dict TypeAdapter The typed lookup only read the eu and us multipliers, so a deployment that set regional_processing_uplift_multiplier_fedramp stopped applying it. /model/update and the model write paths now share one module-level TypeAdapter instead of building one per call. * fix: bill regional uplift multipliers float() accepts, bytes included * fix: raise the in operator's own error from one shared json_contains, and name the bad uplift type |
||
|
tavren27
|
2d2290067c
|
fix(bedrock): run realtime guardrails on Nova Sonic sessions (#49833)
* fix(bedrock): run realtime guardrails on Nova Sonic sessions * fix(bedrock): quote the spoken violation message and skip it after a blocked Nova Sonic tool output * fix(bedrock): screen every message role, end Nova Sonic sessions on blocked transcripts, drop unparseable frames * fix(realtime): apply guardrail masks, keep block reasons out of spoken notices, fail closed on guardrail crashes * fix(realtime): screen joined text parts, keep the documented notice, honor key opt-outs Text parts of one message are joined before screening again, so a phrase split across parts is still blocked, and a mask collapses them into one part. The spoken notice is realtime_violation_message or the block reason, quoted only on Nova Sonic. Client-written output_text and text parts are screened. Key and team guardrail opt-outs reach regular and translation realtime sessions. A non-string upstream transcript skips screening as before. * fix(realtime): honor guardrail role scope, clear stale response.create drops, screen split keywords, report Nova guardrail outages * fix(realtime): keep every per-part mask when a joined text masks differently * fix(realtime): screen one joined text per item and pass the session's tenant context to guardrails Multi-part items reach each guardrail once as the space-joined text, as on main, and only an item a guardrail rewrites gets its parts screened again to keep per-part masks. Guardrails now receive the session request data with its metadata, so key and team policies apply, and a ModifyResponseException block is reported as a block. Nova scopes every message item as the user turn it sends to Bedrock. * fix(realtime): catch phrases split mid-word across text parts without re-screening single-part items * fix(realtime): screen each joined text in its own guardrail call and keep a Nova opening message for the retry while it is screened * fix(realtime): screen text parts in a loop so large messages keep their masks * fix(realtime): fail closed on a non-string transcript under a guardrail and give each turn its own guardrail metadata * fix(realtime): fail a non-string transcript only when a guardrail would screen the user's turn --------- Co-authored-by: mateo-berri <277851410+mateo-berri@users.noreply.github.com> |
||
|
tessellark
|
b5a72c2042
|
fix(router): pin /v1/responses requests for prompt caching affinity (#50321)
* fix(router): pin /v1/responses requests for prompt caching affinity * fix(router): pin previous_response_id turns and keep native tools in the Responses pin key A Responses turn that continues a stored response now goes to the deployment that answered it, unless that deployment already failed this request. Native Responses tools are hashed into the pin key as sent instead of converted, so two requests with different native tools no longer share a pin and a native request no longer logs a false dropped-tool warning. The pin keeps its cacheable deployment ids in a sorted tuple so loggers can serialize request metadata. * fix(router): leave a Responses request with a malformed tool to the provider * fix(router): keep the Responses prompt cache pin out of logged metadata The pre-call check now keeps the pin in a router-owned in-flight store and puts only a random token in litellm_metadata, so loggers that ship metadata never see the routing key or deployment ids. The store is shared across check instances, so turning prompt_caching off and on at runtime keeps pinning, and a retry drops its previous attempt's token. The skipped-input debug log now names only the error type. * fix(router): apply the prompt caching pin after tag, region and pre-call filters, and leave it on retries The prompt caching check no longer narrows the candidates inside the callback filters. It names the deployment it prefers, and the router core applies that preference at the pinned deployment step, after tag, region, pre-call and plugin filters, only when the deployment is still a candidate. A pin can no longer send a request to a deployment its tags do not allow. A retry skips the pin when the pinned deployment was already tried in this request, read from the retry breadcrumbs in both metadata buckets and the excluded and retry-skipped ids. When the retry count outruns the breadcrumbs the check stands down, so a pinned deployment that keeps failing is not retried on every attempt. * test(router): list the prompt caching check as an in-memory cache caller The check's in-flight Responses pins live in a process-local InMemoryCache, so it opens no Redis span and needs no key family, like the other in-memory callers * fix(router): pin a lone healthy deployment, free every in-flight Responses pin, and clear a reused metadata dict's retry records * test(router): lift test cooldowns through the cooldown store's evict_local * fix(router): key Responses tools like their chat twin and drop the pin token after success A Responses function tool now hashes only the function fields it sent, so a chat request and a Responses request with the same tool share one pin. The success event also removes the pin token from the caller's metadata dict. * test(router): cover untyped tool parameters and custom tool grammars in the Responses pin key --------- Co-authored-by: mateo-berri <277851410+mateo-berri@users.noreply.github.com> |
||
|
mrinal-berri
|
aeede53112
|
feat(guardrails): add mcp_block_behavior to raise 400 on MCP tool blocks (#50284)
* Allow guardrails to propagate MCP tool blocks Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * Refresh generated API schemas for MCP block behavior Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * Fix MCP guardrail deployment selection and streaming errors Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * Enable content filter blocks during MCP calls Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * Preserve content filter MCP hook restrictions Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> --------- Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> |
||
|
joshua-berri
|
de0b389205
|
feat(router): add the Switchyard stage router as an auto router strategy (#50407)
* feat(router): add Switchyard Stage Router strategy * feat(router): decision-model classifier and safer fallbacks for the Switchyard stage router Moves to nemo-switchyard 0.4.0a1, which adds decision-model classifiers and makes ModelCall.respond and fail async. The classifier is now a tagged union of an LLM judge and a decision model, and decision calls go through the router's adecisions so any decisions provider works. Algorithm misbehavior and library errors route to the capable model instead of failing the request, and the session id plus client session headers are always passed so capable holds apply without a classifier Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * fix(router): keep the Switchyard router inside the type-check gate The gate's environment did not install the optional switchyard extra, so every libsy symbol counted as Unknown. nemo-switchyard joins proxy-dev like the other optional runtime packages, and imports go through switchyard_rust.libsy, which ships py.typed. Step handling moves into helpers so no Final is rebound in the loop, and the dashboard API types are regenerated for the new routing fields Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * fix(router): route Switchyard on every endpoint and reject unsafe router saves The router now passes each request's call type to routing strategies, and Switchyard converts the request to chat messages with the existing converter for that call type before translating it. /v1/messages turns with tool or thinking blocks used to fall back to the capable model, and now route on their signals. The Anthropic converter can keep tool_result.is_error on an analysis copy, so Switchyard sees tool failures there, while requests sent to models never carry it Saves of a strategy router that lists itself now fail with a 400 instead of a 500 with a dead row, the capable and efficient models must differ, judge and decision calls are not retried before failing open, the Switchyard routing fields are classified for redaction, and the edit dialog returns the saved router instead of the stale one Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * fix(router): catch self-routing on patch and rename, and trim Switchyard tests The self-routing check now reads the stored params through the decrypting helper and runs when a patch carries no litellm_params, so a PATCH that points a router at itself or renames it onto one of its targets returns 400 instead of a 500 with a dead row. The Switchyard tests share one transcript builder and router fixture, and reformatting noise in unrelated test files is reverted Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * refactor(ui): make Switchyard a strategy in the existing auto router form Switchyard is now a card in the existing strategy picker. Picking it swaps the tier and classifier fields for the Switchyard fields, and the shared name, team and save button create it through one submit branch, the same flow as the other strategies. The separate add form, the extracted picker and the hidden-form swap are gone, and the edit dialog keeps its own Switchyard form The judge's session_affinity and message_hash_fallback settings are removed, since Switchyard rejects both for the stage router Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * fix(ui): accept partial tool semantics and show server errors on model edit The Switchyard tool semantics field rejected mappings that list only some categories, although the server accepts them. Model PATCH errors were reduced to a generic message, so it now goes through the shared API client like model create and surfaces the server text Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * feat(ui): show the routing strategy picker up front in the auto router form Every strategy was hidden behind an Advanced configuration link, so the quick setup read as the only option and never said it creates a heuristic router. The picker now always shows with Heuristics as the labelled default, the quick setup sits under it, and picking another strategy opens that strategy prefilled with the chosen models Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * fix(ui): show the server's reason when a model settings or API key update fails Both screens replaced the rejection with a fixed message, so a 400 such as a router routing to itself read only as a generic failure Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * feat(router): route Switchyard on the Rust router core main replaced the Python router core with the Rust one, which left Switchyard routers registered nowhere. Register them with the other pre-routing strategies, hand each strategy the endpoint the request came through, and evict Switchyard routers on reload and delete like the rest Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * fix(router): scope Switchyard sessions per API key and fix review gaps in the Switchyard form Switchyard session state is keyed by a hash of the caller's key hash and session id, and client session headers no longer reach Switchyard unscoped. The add form carries the quick setup models into Switchyard, sends the team and access groups, clears a stale tool semantics error, and the schema reads API-saved configs with nulls and structured decision instructions Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * test(router): make the Switchyard test folder a package * fix(router): read the Switchyard session key hash only from the proxy's own metadata * fix(proxy): keep nested nulls as null when masking deployment params * fix(router): send Switchyard judge calls with the caller's own metadata bucket so tag routing applies * fix(router): put Switchyard judge metadata in the bucket the judge endpoint reads --------- Co-authored-by: Joshua Valluru <326636767+joshua-berri@users.noreply.github.com> Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com> |
||
|
m.torres
|
150f4b706b
|
fix(router): count embedding batches per input in pre-call checks (#49851)
* fix(router): count embedding batches per input in pre-call checks Embedding input lists went through the Responses input transform, which calls .get() on each item, so every list-input /embeddings call logged "failed to count tokens" and skipped the context window check. Lists of texts, token IDs or token ID lists now count as their longest input, since an embedding model's context window applies to each input on its own * fix(router): keep empty and boolean input lists on the Responses count path An empty list carries no embedding evidence, and counting it as a zero token batch skipped the instructions and tools of an empty Responses input. Boolean lists are not token IDs * fix(router): honor disable_token_counter for token ID embedding input Text inputs already counted as zero through token_counter when token counting is disabled, while token ID inputs were still measured by length and could be rejected. The router reads the flag through its dynamic config like its other process settings * refactor(router): type the embedding count helper with a read-only sequence and final locals * fix(router): count embedding batches per input on the media-free text projection The per-input count now runs on the text projection from the embedding media split, so data URIs a deployment embeds as media keep the media path and only text items are checked against the context window. A caption next to an image is checked on its own, and a text-only deployment still counts an image data URI as text * fix(router): keep embedding images out of the pre-call count only on image backends, and skip tokenizing inputs that fit by bytes Media leaves the embedding count only when every deployment in the group declares image embedding input, so a text-only backend still counts a data URI as text and an image backend outside the TPM media list no longer gets a false 400. A text no longer than the smallest window in UTF-8 bytes counts as its byte length without the tokenizer, since a token is never shorter than a byte. * fix(router): keep embedding images out of the pre-call count when any backend in the group embeds images Mark Cohere embed-english-light-v3.0 as embedding images, as its v3 siblings are. * fix(router): check embedding text batches only with an exact OpenAI tokenizer litellm's default tokenizer counts non-English text several times higher than non-OpenAI embedding providers do, so counting batches with it refused inputs the provider accepts. Text batches on those groups now stay unchecked as at the merge base, without the error log. Main's per-provider media split and the cost map are restored, so strings and embedding media keep main's path. * fix(router): count long embedding batches in slices so chat counts do not wait behind them A text batch is now tokenized in slices of 65,536 characters, each its own offloaded call, so the FIFO count limiter can serve a chat count between slices. The no-OpenAI-tokenizer rule for text lists moved into the pre-call check, so RoutingContext.count_pre_call_check_tokens counts a text list by its longest text. * fix(router): read embedding media off the event loop and wait on the count limiter in the fairness test --------- Co-authored-by: mateo-berri <277851410+mateo-berri@users.noreply.github.com> |
||
|
simonp9
|
1f15088fc9
|
feat(proxy): mixed Codex model picker and LiteLLM routing on native codex routes (#49408)
* feat(proxy): mixed Codex model picker and LiteLLM routing on native codex routes
The native Codex catalog now lists the key's other chat and responses models after
ChatGPT's, within Codex's catalog size limit, and drops the ETag so the merged list
refreshes on Codex's timer. A request for a model configured by exact name on another
provider goes through /v1/responses without the ChatGPT bearer or account header,
while ChatGPT names and wildcard-only matches keep the subscription passthrough.
Reasoning items another provider produced are dropped before forwarding to ChatGPT.
* fix(proxy): keep Codex's picker from turning reasoning off for LiteLLM models
Codex fills a missing default_reasoning_level with effort "none" when a model is picked in /model, while
`codex -m <model>` sends no effort at all. A thinking-only provider model then answers 400. On the native
/backend-api/codex/responses route, drop effort "none" for a LiteLLM model whose catalog entry names no
default, so the provider applies its own default exactly as `codex -m` does
* fix(proxy): harden Codex mixed-model routing on the native ChatGPT routes
Keep the ChatGPT credentials out of hook raw headers on the inner /v1/responses
request, route to LiteLLM and merge LiteLLM catalog rows only for keys that may
call /v1/responses, drop unpersisted reasoning ids on the passthrough when store
is not true, keep any slug the caller's ChatGPT account lists on the subscription
via a shared account catalog cache, and keep X-Models-Etag for keys whose catalog
added no LiteLLM rows. The guide covers the new behavior and the /v1 wildcard
precedence over a catch-all deployment.
* fix(proxy): keep Codex's native route on the subscription when the account list is unknown
A name OpenAI lists for the caller's ChatGPT account now goes to LiteLLM only when the
account's list is known and leaves it out. A failed catalog read is remembered for 60 s and
the Codex version comes only from a User-Agent whose product matches the originator header.
The merged catalog serves a stable ETag digest of OpenAI's ETag and the key's LiteLLM rows,
and /responses replaces X-Models-Etag with the same value, memoized per key for 300 s, so
Codex stops refetching or dropping its catalog cache.
LiteLLM-routed native replies wrap their reasoning with the model id, so an OpenAI or Azure
blob is dropped before chatgpt.com and unwrapped for LiteLLM, and the chat-completions bridge
decodes wrapped thinking blocks with their signature.
The access-group recipe uses chatgpt/ model names so /v1 routing does not change, and the
native route checks the chatgpt/ form of a name the key holds only through that group. A
none reasoning effort is dropped only under drop_params. The guide covers app tools, every
routing case, and the ETag, and the catalog logs how many models the byte limit left out.
* fix(fireworks_ai): move Codex additional_tools items into tools on the Responses API
* fix(proxy): keep Codex native turns working when the key's model listing fails
A listing callback that raises no longer turns chatgpt.com's catalog or
responses answer into a 500. The catalog keeps upstream's body and ETag, and
responses keep upstream's X-Models-Etag.
Under drop_params, a `none` reasoning effort is now dropped only when a
candidate deployment does not explicitly support it, so an alias to a model
like openai/gpt-5.4 keeps it.
The account slug list and its unavailable marker are keyed on the account id
plus the caller's ChatGPT Authorization header, so a failed fetch with one
bearer cannot turn off LiteLLM routing for another caller on the same account.
The auth step checks the Codex route before importing the Codex module.
* fix(proxy): keep Codex model switches working across LiteLLM providers
On the Codex native route, a LiteLLM model now gets only the encrypted reasoning every deployment it can route to is able to decrypt, so ChatGPT or another deployment's blobs no longer make xAI answer 400
Codex additional_tools items are hoisted into tools for every provider except the OpenAI API, and namespace tools become flat tools named mcp__docs__search, the shape xAI accepts
A JWT caller's models ETag is cached on a hash of its LiteLLM credential, so ChatGPT turns stop rebuilding the catalog rows
* fix(proxy): keep Codex namespace tool calls intact across LiteLLM providers
Fireworks and Bedrock Mantle accept Codex namespace tools natively, so the
native route now hoists them without flattening, like OpenAI and Azure and
the chat completions bridge. Other native Responses providers still get
flat names, now built with the bridge's namespace_tool_name_map, and each
function_call they return gets its tool name and namespace back in the
non-streaming body and in streamed output_item.added, output_item.done,
and completed/incomplete/failed events. Every other event passes through
unchanged
* fix(proxy): bridge Codex turns for providers whose Responses API rejects Codex tools
Codex sends its freeform exec tool as a custom tool on every turn, and xAI's
Responses API answers 422 for custom, namespace and additional_tools shapes.
On the Codex native route, a model whose deployments include a native
Responses provider other than OpenAI, Azure, Fireworks or Bedrock Mantle now
asks for the chat completions bridge with use_chat_completions_api, which
already turns custom and namespace tools into functions, hoists
additional_tools, and puts the namespace back on each returned call
OpenAI and Azure still get the request unchanged, and Fireworks and Bedrock
Mantle still get it on their own Responses API with additional_tools hoisted.
The route's namespace flattening and the response-side namespace restore are
removed, since no path sends flattened names anymore
* fix(proxy): drop reasoning ids a Codex turn's LiteLLM target cannot resolve
A reasoning item another deployment produced, replayed with store false and no blob the
target can decrypt, now reaches the target as its readable summary without id or status,
and is dropped when nothing readable is left. OpenAI answered 400 on the status field and
404 on the id for a glm-5p3 item replayed whole to an openai deployment. Same-deployment
replays and stored conversations keep the item whole
* refactor(codex): reach encrypted reasoning stripping through the router registry
The Codex route read Router.optional_callbacks and imported EncryptedContentAffinityCheck from
litellm.router_utils, which the router surface fence forbids outside the router. RouterRegistry
gains decryptable_reasoning(model_group, request_input, targets), which keeps the same rule: with
encrypted content affinity on for the group, reasoning one target can decrypt stays, otherwise
every target must decrypt it
* fix(proxy): keep Codex's priority tier off LiteLLM rows in the native model catalog
* fix(proxy): drop Codex's none effort for an alias whose native entry names no default
* fix(proxy): start a native Codex alias at its pinned reasoning effort and build the catalog off the event loop
* test(proxy): map the pinned reasoning efforts field in the router registry stub
* fix(proxy): honor admin-only /v1/responses and keep Codex's default tier off native aliases
Codex native routes now send a key to LiteLLM only when admin_only_routes allows it,
a native alias with configured service tiers starts with no tier picked, and the
account slug cache declares its Redis key family.
* fix(proxy): check the key's tool allowlist on native Codex LiteLLM turns and additional_tools items
* fix(proxy): check vector store grants on file_search tools inside additional_tools items
A Responses request can define file_search inside an additional_tools input item, which the native Codex route hoists for LiteLLM-served models and OpenAI runs as is on /v1/responses. The vector store check now reads those tools in both strict and legacy modes. The input item scan uses a Mapping guard instead of a pydantic error per item, since auth now runs it on every list input.
* fix(proxy): apply the proxy-wide allowed_routes to LiteLLM models on native Codex routes
* fix(proxy): cap native Codex account catalog reads per LiteLLM caller before rate limits apply
* fix(guardrails): check additional_tools items next to top-level function tools in tool_policy
ToolPolicyGuardrail read tools from the request body only when no top-level
function tool existed, so a blocked tool defined in an additional_tools item
reached Fireworks once the Responses transform moved the item into tools after
the pre-call hooks. Check every list on each request, including namespace
members inside items.
* fix(router): serve decryptable_reasoning from the Rust core's optional checks
The merge with main removed the Python router core, so the registry's host has no
optional_callbacks and no routing context, and native Codex requests with encrypted
reasoning raised RustCoreUnsupported. The filter now lives with the optional checks,
which own both. The ported pinned_reasoning_efforts field gets a test on the Rust core
Also give the static litellm.Router a declared constructor. Since the Python core was
deleted a bare __init__ assignment reads as RustCore's, so every Router(model_list=...)
is a type error and make lint fails on main. TYPE_CHECKING only, no runtime change
* fix(proxy): check the budget of a saved alias target on native Codex LiteLLM turns
Native /backend-api/codex/responses binds {endpoint}, so auth skipped the key's or team's model_group_alias and read budgets on the requested name, while the later /v1/responses handoff served the target unchecked. Auth now resolves the alias on a turn that routes to LiteLLM, as it does on /v1, and leaves ChatGPT-only names on the subscription. Also makes the tool_policy tests build a real ToolPolicyRegistry and type their fixtures.
* fix(proxy): keep a name the ChatGPT account lists on the subscription despite a saved alias
* fix(codex): keep LiteLLM keys off ChatGPT, enforce tool policy on namespaces and pin reasoning on aliases
Auth no longer asks ChatGPT for the account's model list when the bearer is
missing or is the LiteLLM key, so a key sent as the bearer never reaches it.
tool_policy now reads the members of a top-level namespace, which is where
the native route's hoist puts them.
A global alias now reads encrypted content affinity from its target group.
A non-list top-level tools value and a non-string function name now reach the
provider, which refuses them, instead of raising a 500.
The catalog ETag no longer depends on the order a process iterates priced
service tiers in, so workers and pods agree on it.
* fix(proxy): list a team's models in the order the team set them
get_team_models built the names in a set, so each process listed a team key's models in its own hash order and the native Codex catalog sent a different row order and ETag from each worker
* fix(proxy): apply JWT route policy before native Codex routes serve LiteLLM models
A standard JWT caller now gets a LiteLLM model on /backend-api/codex/models
and /backend-api/codex/responses only when the JWT auth flow would admit a
direct /v1/responses call for that model, so team_allowed_routes,
admin_allowed_routes and RBAC route limits hold on the handoff too.
The Fireworks Responses transform also leaves additional_tools input items
in place when the tools param is not a list, instead of raising while hoisting
* fix(fireworks): hoist additional_tools when tools is a tuple
litellm.responses passes a tuple tools argument through to the Fireworks
transform unchanged, so the non-list guard skipped hoisting and left the
additional_tools item in the input. Treat tuples like lists and cover both
shapes in the drop_params merge test
---------
Co-authored-by: mateo-berri <277851410+mateo-berri@users.noreply.github.com>
|
||
|
ryan-crabbe-berri
|
ec46a981fc
|
feat(model-hub): show each deployment's full prices in the model hub (#50145)
* feat(model-hub): show each deployment's full prices in the model hub The model hub only showed one input and one output price per model, the highest across its deployments. Each model group now lists every deployment's prices, read from the same entry billing charges, including cached tokens, long-context tiers, flex, priority, batch and tiered ranges. Both the logged-in and public hub detail views render them in one table per deployment, labeled by provider only * refactor(ui): move model hub pricing breakdown text onto i18n Route every string the deployment pricing breakdown writes through t() under a new deploymentPricing group in en.json, plus the three hub headings that host it. The English on screen is unchanged * fix(model-hub): show Vertex regional prices and explicit cached reads as billed A Vertex deployment outside the global endpoint now shows its token prices with the regional uplift billing applies at the deployment's location. The explicit cached read price shows per 1M tokens on its own row instead of a raw per-token value. The breakdown also says that adjustments made per request, like off-peak discounts and regional or data residency surcharges, aren't included * fix(model-hub): show prices written as text the way billing reads them A price written in config.yaml as 3e-7 arrives as text. Billing turns it into a number, but the hub dropped it, so a deployment showed no price or the cost map's price instead of its own. The hub now reads numeric text for regular prices and size-based price ranges, and skips text that isn't a number * style(ui): format public model hub test * test(model-hub): name the deployment pricing fixture to stay within the inline object lint budget |
||
|
Kerry Lu
|
d4f338ef1e
|
feat(tags): reset a tag's spend through /tag/update (#50323)
* feat(tags): reset a tag's spend through /tag/update Co-authored-by: Arnold Galovics <arnold_galovics@docktape.com> * test(tags): assert spend update response contract * fix(tags): seed spend counter after evicting tag cache * refactor(spend): expose shared counter reset helper * fix(spend): sanitize counter key in spend counter warning log --------- Co-authored-by: devin-ai-integration[bot] <158243242+devin-ai-integration[bot]@users.noreply.github.com> Co-authored-by: Arnold Galovics <arnold_galovics@docktape.com> |
||
|
mrinal-berri
|
bc1309f18e
|
fix(responses): map streamed flex_unavailable errors to 429 not 500 (#50336)
* fix(responses): map streamed flex_unavailable errors to 429 not 500 When OpenAI has no flex capacity, a service_tier flex request fails with type resource_unavailable / code flex_unavailable. Non-streamed, this arrives as HTTP 429 and litellm raises RateLimitError. Streamed /v1/responses delivers the same error as an in-stream event, and neither field was in _ERROR_CODE_HTTP_STATUS, so it defaulted to 500 InternalServerError. The HTTP status then depended only on stream: one provider outage looked like a 429 or a 500. The 500 reads as a server error in metrics and alerting, and callers that retry on 429 do not retry it. Map both resource_unavailable and flex_unavailable to 429 so the streamed path matches the non-streamed path. Fixes #49736 * test(responses): cover streamed flex outage end to end for error and response.failed events Streams the flex_unavailable payload from #49736 through litellm.aresponses with an injected client and asserts a 429 RateLimitError for both in-stream event kinds, matching the non-streamed path * test(responses): drop explanatory comment from flex status mapping test * test(integration): cover streamed flex outage status on responses and chat bridge Drives the flex_unavailable payload from #49736 through a real proxy as an in-stream error event and as response.failed. The native /v1/responses stream fails with rate_limit_exceeded and logs 429, and the chat bridge answers a 429 throttling error, where both reported a 500 before the mapping fix --------- Co-authored-by: BinYi LIU <bliu1@berkeley.edu> |
||
|
Mateo Wang
|
36c64ce85d
|
feat(bedrock): default to native structured output on Opus 5.5, Fable 5.1, Sonnet 5 and Haiku 5.5 (#50221)
* feat(bedrock): enable native structured output by default for selected Claude models Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * fix(bedrock): keep govcloud and bare model rows off native output Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * fix(bedrock): retry rejected native schemas with the JSON tool on default-native models Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * fix(bedrock): keep the rejection retry on the unforceable-tool native path Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * fix(bedrock): retry rejected schemas on the Messages path for default-native models Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * style(bedrock): format Messages native output retry Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(bedrock): allow profile-only native output metadata Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * fix(bedrock): only disable default-native output on schema-rejection retries Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(bedrock): type fixture parameters in native output retry tests Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * fix(bedrock): retry with the JSON tool when Bedrock rejects a schema's grammar size Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * fix(bedrock): drop strict tools after a native-output retry hits the grammar limit Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * style(bedrock): format native output fallback Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * fix(bedrock): honor the deployment native output opt-out and keep wide batch schemas on the JSON tool A deployment's supports_bedrock_native_structured_output: false now keeps the JSON tool on rows that default to native output. Bedrock batch has no retry, so a record whose schema has more than 64 properties keeps the JSON tool there too. * fix(bedrock): retry a native structured output stream as a buffered stream when Bedrock denies the stream action A role allowed only bedrock:InvokeModel got a 403 on schema streams once these models went native, since native streams call converse-stream while the JSON tool path buffered through converse. The retry now sends that request again as a buffered JSON tool stream, the shape the base sent. The wire test moves its neither-key and opt-in cells to Opus 4.7, its Bedrock-key-only cells to Sonnet 5.5, and its unforceable-tool cells to the bare Opus 5.5 id, since this PR makes Opus 5.5 and Haiku 5.5 always native. The batch wire test now checks a thinking schema record on both sides of the contract the batch unit test already pins, the JSON tool on opt-in Sonnet 5.5 and native output on always-native Haiku 5.5, since its Haiku 5.5 and Opus 4.8 cells assumed those models were opt-in. * fix(bedrock): move the Claude 5 native output default to a key released proxies do not read Released proxies load the cost map from main and have none of the native output retries, so setting supports_native_structured_output on the Opus 5.5, Fable 5.1, Sonnet 5 and Haiku 5.5 profile rows turned working wide or min/max schemas into 400s on v1.108.1. Those rows keep supports_native_structured_output false and gain bedrock_native_structured_output_by_default true, which get_model_info reports as supports_native_structured_output, so this version behaves as before while released versions keep the JSON tool. * fix(bedrock): rename the Claude 5 native output default key to supports_bedrock_native_structured_output_by_default The cost map guard runs main's model catalog on the PR's cost map, and that catalog rejects any undeclared key that does not start with supports_. Declaring the old name on main first needs its own PR, so the key takes the supports_ prefix. Released proxies still never read it * fix(bedrock): honor registered and batch opt-outs, move Opus 5 and 4.8 to the default key A registered supports_bedrock_native_structured_output false now beats the native output default, and batch uploads read the deployment's model_info, so an opted-out deployment keeps the JSON tool on every record. The Opus 5 and Opus 4.8 profile rows move to supports_bedrock_native_structured_output_by_default, so released proxies on main's cost map keep the JSON tool on them. The 64 property batch cutoff now applies only to rows with that key, so models that were already native keep their batch output shape. * fix(proxy): report a deployment's Bedrock native output opt-out on /model/info /model/info merged the deployment's model_info over the backend row, so a deployment that set only supports_bedrock_native_structured_output: false still showed the backend's derived supports_native_structured_output: true. The generic flag is now derived from the merged info unless the deployment set it itself, through the same helper get_model_info uses. * fix(bedrock): keep native output when a denied schema stream is resent buffered A role without bedrock:InvokeModelWithResponseStream had its native output schema stream resent as a buffered JSON tool call. The resend now keeps the native params with fake_stream set, and falls back to the JSON tool only if Bedrock then rejects native output. * fix(bedrock): drop rejected strict tools on the JSON tool retry after a buffered native resend * fix(bedrock): send a buffered schema stream to /invoke and keep the by-default key on its deployment The shared HTTP handler now builds the non-streaming URL for a fake-streamed request, so the invoke route's buffered resend after a denied stream calls /invoke instead of invoke-with-response-stream. supports_bedrock_native_structured_output_by_default is deployment scoped, so setting it on one deployment no longer reaches sibling deployments of the same model * fix(bedrock): answer a buffered invoke stream from the non-streaming reply * fix(bedrock): keep the JSON tool retry buffered after a denied stream's schema is rejected * fix(bedrock): count only a true fake_stream and keep the deployment by-default key out of /model/info A fake_stream sent as "true" or 1 picked the non-streaming URL but kept the streaming reader, so Bedrock invoke streams answered 500 after three billed calls. Only a real True counts now, as on main. /model/info read supports_bedrock_native_structured_output_by_default from the deployment, while routing reads it only from the cost map, so it reported native output for deployments that use the JSON tool. --------- Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> |
||
|
Kerry Lu
|
ddd162439f
|
fix(proxy): stop /tag/update 500 when the body carries the tag's budget_id (#50324)
Co-authored-by: devin-ai-integration[bot] <158243242+devin-ai-integration[bot]@users.noreply.github.com> |
||
|
Alexander Chernov
|
d1cde1002e
|
fix(chatgpt): forward text and max_output_tokens to the Codex backend (#38086)
* fix(chatgpt): keep text and accept string input for the Codex backend
Two request-side defects stop chatgpt/* serving structured output or a string
input.
The allowlist at the end of transform_responses_api_request omits "text". The
chat-to-responses bridge translates response_format into text.format, so the
allowlist discards strict schemas silently and the backend answers with prose
instead of JSON. The backend does honour text.format when it receives it.
Separately the backend rejects a bare string input with
{"detail": "Input must be a list"}, while the Responses API itself accepts
either shape, so the string needs wrapping before it is sent.
Both are request-shaping only, so neither depends on the streaming work in
#31332 or #34095, though those two block the same call and have to be in place
to observe this one end to end.
The "text" case was reported in #24356 and closed by the stale bot without a
fix.
Signed-off-by: Alexander Chernov <alexander@chernov.it>
* fix(chatgpt): annotate the input coercion for the type-discipline gate
The LIT002 budget counts mutable-collection construction, and wrapping the
string input builds a list and a dict, which put the total two over its
ceiling. Both constructions are the wire payload this backend requires, so the
list cannot be replaced with an immutable equivalent; annotate the line with a
reason instead, matching the convention used elsewhere in the tree.
LIT005 freezes reasonless suppressions at zero, so the reason is required
rather than decorative.
Signed-off-by: Alexander Chernov <alexander@chernov.it>
* refactor(chatgpt): bind the coerced input instead of rebinding the parameter
Greptile flagged the parameter rebinding and the explanatory comments against
the repo conventions in CLAUDE.md, which bans rebinding a function parameter
(LIT011) and keeps comments to suppressions, TODOs and genuinely complex logic
Bind a Final local for the coerced input rather than reassigning `input`, and
drop both prose comments; the rationale lives in the commit and the PR body
rather than duplicated at the call site. The mutable-ok suppression stays,
since it is the allowed kind and the list is still constructed
Behaviour is unchanged: 24 tests pass, and the same 4 fail with only the
provider file reverted to base
Signed-off-by: Alexander Chernov <alexander@chernov.it>
* fix(chatgpt): forward max_output_tokens to the Codex backend and return incomplete non-stream responses
The Codex backend accepts and enforces max_output_tokens, so the allowlist now keeps it
instead of dropping it. A structured-output request could otherwise stream until the
backend gave up, ignoring the caller's max_tokens.
A capped request ends with response.incomplete, so the non-stream SSE aggregator now
treats that event as terminal and returns the partial response instead of an error.
---------
Signed-off-by: Alexander Chernov <alexander@chernov.it>
Co-authored-by: mateo-berri <277851410+mateo-berri@users.noreply.github.com>
|
||
|
JeongCheolLee
|
febbe38326
|
fix(anthropic)!: do not surface server_tool_use as client tool_calls (#48385)
* fix(anthropic): do not surface server_tool_use as client tool_calls Anthropic server-executed tools (web search, tool search, code execution) were returned in chat completions tool_calls with srvtoolu_ ids, so OpenAI-protocol agent clients treated them as pending work and re-invoked the model until their recursion limit. Move them to provider_specific_fields.server_tool_calls on both the non-streaming and streaming paths. History replay, the /v1/responses bridge and the /v1/messages streaming guardrail read that field, so multi-turn reconstruction, web_search_call output items and the guardrail's fail-closed check on server tool blocks keep working. * fix(responses): return Claude web_fetch calls as web_search_call open_page items * fix(responses): keep Claude server web calls without a result as hosted items A non-streamed Claude turn that calls web_search or web_fetch next to a client tool can return the server call before its result. Build an in_progress web_search_call for it, so the Responses bridge no longer returns it as a client function_call * fix(anthropic): keep pending server calls native and hide server calls from client tool paths Replaying a server call that has no result yet now sends it back as server_tool_use, so Anthropic accepts the next turn. The tool permission guardrail checks server calls again and rewrite mode removes a denied one. The /v1/messages adapter keeps Claude's server blocks, and the Responses bridge never returns a server call as a client function_call. * fix(anthropic): let guardrails, tool usage, tracing, and Responses replay see Claude's server calls Claude's server tool calls moved out of tool_calls into provider_specific_fields.server_tool_calls, so every reader of tool_calls stopped seeing them. The chat guardrail handler now passes them to output guardrails and writes rewrites back to server_tool_calls (stream and non-stream), tool usage and registry discovery count them, otel, Datadog and the Logs Tools panel list them, and a pending Claude web search or fetch item replayed to /v1/responses goes back to Anthropic as its own server_tool_use instead of running again * fix(anthropic): let guardrails and loggers read Claude's server tool calls again Guardrails, tool policies, tracing loggers and the Responses guardrail translation read only client tool calls after server calls moved to provider_specific_fields. They now read server calls first, then client calls, through one shared helper, and the Responses guardrail treats Claude's web_search_call items as web_search or web_fetch calls. Non-streaming splits server calls by block type, as streaming does. * fix(anthropic): keep Claude server calls visible to guardrails, redaction and tracing A guardrail rewrite of a Claude server call now also updates its hosted web_search_calls copy, streamed or not. Redaction with message logging off hides server call arguments and hosted web call actions on chat and Responses. Pillar and Pangea scan server calls, transform-mode streaming guardrails keep server call chunks, and OTEL v2 Responses spans list Claude's web calls. * fix(anthropic): keep the legacy hook view of streamed /v1/messages web search after the main merge Main's legacy post-call hooks see Claude's server call as tool_use with its search results first. The streamed /v1/messages response now keeps that view while native scans keep server_tool_use. * fix(guardrails): let Cisco AI Defense scan Claude's web calls on Responses Claude's web search and fetch now come back from /v1/responses as web_search_call items, so Cisco AI Defense reads and clears their query or URL the way it did for the old function_call arguments. The helpers that read and rewrite those items move to the shared guardrail utils so the Responses guardrail handler and Cisco use one copy. * test(guardrails): inject Pillar's HTTP client and type the Cisco web call actions * test(guardrails): type the fake Pillar client like its Protocol * fix(anthropic): keep a pending server call and its thinking when the turn also waits on a client tool Only a turn with no client tool_use is a paused turn. A tool_use turn that also started web_fetch or web_search still needs that call and its thinking replayed, or Claude loses the fetch result * fix(guardrails): scan replayed Claude server calls and hosted web calls Request guardrails now see and rewrite server_tool_calls replayed in chat messages. Cato and Purview scan Claude web_search and web_fetch calls on Responses output, and a function tool call with an extra custom key is no longer dropped from Cisco, Purview, Repello and PANW scans. * fix(guardrails): scan replayed Claude server tool calls in Lasso, Lakera and Qualifire request checks * fix(anthropic): apply streamed guardrail tool call rewrites to their own block when a server tool call follows * test: annotate the new server tool call test locals with Final and type their parameters * test: type the new guardrail test helpers with litellm's tool call types instead of Any * test: narrow the remaining new test helper parameters from bare dict and object * fix(anthropic): keep server calls in Responses streams, guardrail masks, OTEL and Logs Streamed /v1/responses now emits Claude's tool search and paused web search calls, text guardrail masks rewrite replayed server calls, OTEL gen_ai messages and the Logs pretty view list server calls next to client calls, and integration tests read server calls from server_tool_calls * fix(guardrails): keep hosted web call copies in sync on request masks and denied server calls, and accept null tool_calls in Lakera * fix(anthropic): serialize masked hosted web call copies through their adapter so a replay before the OpenAI model is built does not 500 --------- Co-authored-by: mateo-berri <277851410+mateo-berri@users.noreply.github.com> |
||
|
Siddharth
|
2a4f8b1d19
|
fix(bedrock): drop another model's signed thinking when replaying to a non-Claude Converse target (#50073)
* fix(bedrock/converse): omit reasoning signature for non-Anthropic models (#49969) * test(bedrock): cover signed reasoning replay in mapped tests * fix(bedrock): drop another model's signed thinking when replaying to a non-Claude Converse target A signed thinking block from Claude replayed to a Converse model known not to be Claude, such as GPT-6 Luna or Nova 2, now leaves the request instead of going out as reasoningContent. Bedrock rejects that field on those models, so the whole request answered 400 A target that may still be Claude, such as an application inference profile ARN with no base_model, keeps the signed block, since Claude needs it on a tool-use continuation. Unsigned blocks keep the existing text path The thinking-block error matcher added earlier on this branch is removed, because Converse and non-Claude Bedrock requests never reach that retry * fix(bedrock): keep signed thinking for Claude behind a region prefix litellm does not list ca. and in. inference profiles name Claude after a region prefix get_bedrock_cross_region_inference_regions does not have yet, so they counted as non-Claude and lost their signed thinking. The thinking-only turn test now pins the exact merged and continue-message shapes. * fix(bedrock): limit the region-prefix Claude match to signed thinking and drop a dropped block's cache point * fix(bedrock): drop Claude's bridged reasoning items before a Bedrock Responses API replay * fix(bedrock_mantle): drop litellm-minted reasoning items before a Mantle Responses call * test(integration): expect Bedrock and Bedrock Mantle Responses to drop minted reasoning Bedrock and Bedrock Mantle answer 400 on a reasoning item litellm minted, so the wire test now expects both to drop it, like OpenAI and Azure --------- Co-authored-by: mateo-berri <277851410+mateo-berri@users.noreply.github.com> |
||
|
moe-berri
|
b02bb99b08
|
feat: add OpenRouter-style -latest model aliases for first-party providers (#50315)
* Add first-party floating model aliases Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * Align Moonshot alias parameter test with its target Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * Share floating alias proxy budgets and rate limits with targets Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * Prove target-only team budgets count and block floating aliases Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * Share floating alias quotas with project and organization targets Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * Coalesce colliding floating alias rate limit counters Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * Fix latest alias Rust fallback and failure accounting Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test: adapt latest alias checks to typed rate-limit store Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> --------- Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> |
||
|
ishaan-j-berri
|
80194a7d70
|
fix(anthropic): preserve Azure Responses reasoning replay (#50024) (#50381)
Fixes #50024 Preserve encoded reasoning history through the existing Azure Responses bridge. |