* fix(security): bump grpcio and boto3 so their bundled CA files drop the compromised eMudhra roots
grpc ignores SSL_CERT_FILE and verifies every channel against its own bundled roots.pem. In 1.78.0 that file still trusts emSign ECC Root CA - G3 and AAA Certificate Services, so Vertex AI, Google KMS, Chirp STT and OTLP gRPC exporters (Arize) accept all 14 advisory chains. 1.84.1 is the first release whose roots.pem drops both roots.
botocore uses certifi when it is installed and falls back to its bundled cacert.pem otherwise. Through 1.43.116 that file still trusts AAA Certificate Services. Requiring boto3 1.43.117 pulls in botocore 1.43.117, the first release whose cacert.pem rejects all 14 advisory chains, so the fallback stays safe if certifi is ever absent
* fix(docker): provide OpenSSL 3 when generating Prisma engines
The Admin UI upgrade notice read GitHub's latest release, which the release
pipeline sets on its own, so admins were offered releases the changelog had
not announced yet and sent to the auto-generated GitHub release page. The
notice now reads the newest stable release from the docs changelog feed and
links to its changelog page. GitHub release notes are still used for the
feature and fix counts of that version's minor line.
Co-authored-by: yucheng-berri <293926326+yucheng-berri@users.noreply.github.com>
* test(interactions): assert the live status enum is parsable instead of pinning it
* test(interactions): read the spec from a vendored snapshot and canary the live one daily
* test(interactions): run the canary on PRs touching its inputs, file an issue on a red run, pin the regenerate command
* test(interactions): take the live spec URL as a pytest option, assert its source, split the canary's tracking-issue job
* test(interactions): file the canary issue whenever its compliance step did not pass, match the issue title exactly, cover UrlContextResult
Update certifi, Rust trust roots, and the Docker CA bundle to remove emSign ECC Root CA - G3
Require the fixed certifi release in litellm-core and retain libssl3 in Docker runtimes for Prisma compatibility
Co-authored-by: shin-berri <shin-laptop@berri.ai>
Every other crate names its error Error. The format crates already re-exported
it under that name, so only the inference crate and a few importers change
* refactor(types): replace Any with proven types in 100 files
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* refactor(types): keep untyped parsing where strict validation changed accepted shapes
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* refactor(types): keep the OpenAI-compatible response parse untyped
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* refactor(types): keep only annotation changes, drop new runtime validation
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>
* fix(responses): round-trip Anthropic compaction blocks as compaction items
/v1/responses on Anthropic routes now returns Claude's compaction block as a
compaction output item (and streams it as output_item.added/done), and a
compaction input item is replayed to Claude as its native compaction block,
so the next turn reuses the summary instead of compacting again. Other routes
receive the summary as assistant text. The Anthropic stream handler now keeps
the summary text that arrives in compaction_delta.
Fixes#41456
* fix(responses): replay compaction blocks natively only when the request compacts
* fix(responses): carry compaction through dict-form edits, SDK replay, affinity, and guardrails
Read Anthropic's dict-form context_management, accept SDK ResponseCompactionItem
objects on replay, strip the encrypted_content affinity wrap before decoding,
scan and rewrite compaction summaries in Responses guardrails, drop the history
a summary replay replaces, send the summary as a user message so thinking
ordering holds, and send the streamed compaction .done with the newest block.
* fix(responses): count, route, guard, and redact compaction items consistently
A compaction item now reads as its summary everywhere the proxy counts or routes
a Responses request: pre-call context checks, the TPM limiter, the input token
endpoint, the stream usage estimate, and the previous_response_id session rebuild
all drop the history the item replaces. A Claude compaction item sent to a
non-Anthropic model (including a fallback) becomes a user summary message instead
of an encrypted blob the provider rejects, and Claude on Bedrock Claude Platform
takes the summary path. Output guardrails see the summary of a returned
compaction item, and logging redaction hides it.
* fix(responses): start previous_response_id replay at a stored compaction and count reservations from it
* fix(responses): keep a stored compaction summary whole in spend logs
* fix(responses): keep stored compaction items summary-only, redact compaction_delta, and finish compaction before reasoning
A spend-log row now stores a top-level compaction input item as its summary alone, so extra keys in a
client-built item no longer reach the database whole. Message redaction also nulls the streamed
compaction_delta, and the streaming bridge sends the compaction item's done event before the reasoning
item that follows it. Adds the audit's integration tests for compaction items on the wire.
* test(responses): cover stored compaction items, compaction_delta redaction, and compaction done order
Adds integration cells for the summary-only spend row of a caller-sent compaction item and its previous_response_id replay, a redacted chat stream row that holds no streamed compaction_delta, and a compaction then thinking stream whose done events follow output order.
* fix(spend_logs): honor the redaction header on failure rows for litellm_metadata routes
A failed /v1/responses or /v1/messages call keeps its proxy request headers in
litellm_metadata, but the failure row read redaction headers from metadata only,
so x-litellm-enable-message-redaction never redacted those rows.
* fix(spend_logs): lift only the redaction opt-in header into failure rows
The failure hook on litellm_metadata routes copied every real request header into the caller's own metadata dict, so the stored request body gained the headers, a caller's metadata.headers was replaced, and a disable header loosened global redaction on failed calls. It now copies only the enable headers into a new litellm_params metadata dict, and only when the caller sent no metadata.headers, so no failure row is less redacted than on main
* test(responses): cover failure-row redaction headers and caller metadata on litellm_metadata routes
* fix(responses): keep earlier-turn instructions after compaction and count compaction items by their summary
A previous_response_id rebuild that starts at a compaction now keeps the system and developer
messages of every earlier turn. Token estimates for raw Responses input count a compaction item
as its summary text instead of the serialized item.
* fix(responses): drop the stored history a caller-sent compaction item replaces on previous_response_id
* fix(responses): count only message items with a string role as compaction instructions
* test(responses): flatten the last two-for comprehension in the compaction wire test
* fix(responses): replay from a compaction in stored Responses output and SDK compaction items
A previous_response_id whose stored row is a Responses output holding a compaction item now starts its replay at that compaction, and the handler dumps SDK compaction objects before checking for a summary. Adds tests for the uncovered compaction paths.
* fix(responses): keep the stored tool call a function_call_output answers when a compaction item replaces history
* fix(responses): keep a stored tool call on every turn after a compaction item and bound the stored summary
A previous_response_id rebuild that starts at a turn holding a compaction item now sends again each stored call a
replayed tool output answers, so a retry or a later turn no longer loses the turn-1 call. The start is picked only from
rows up to the referenced response. A spend row keeps a compaction summary up to the larger of 64K chars and
MAX_STRING_LENGTH_PROMPT_IN_DB instead of with no bound.
* fix(responses): replay a compacted session only through the referenced response
A later sibling turn that sent a compaction item put its summary between a stored tool call and its tool output, so Claude answered 400 on a turn that continued the first of two retried turns, and on a retry sent without the item
* fix(spend-logs): keep main's stored value for a prompt length limit under 2
At a MAX_STRING_LENGTH_PROMPT_IN_DB of 0 or 1 the shared truncation helper stored only the truncation marker, where
main stores the marker followed by the whole value
* fix(spend-logs): keep main's stored tail for a negative prompt length limit
At a negative MAX_STRING_LENGTH_PROMPT_IN_DB the shared truncation helper stored one more character of each value than main, since it dropped main's clamp of the end slice to the limit minus the start
* fix(responses): keep a compaction item when a guardrail rewrites its summary through structured_messages
* test(responses): run the compaction wire tests on the worker database under isolated workers
* fix(router): keep the encrypted content affinity pin out of provider requests
* fix(proxy): inspect compaction summaries in guardrails, strip a client-sent affinity pin, and price failures from the newest compaction
* fix(proxy): count a failed chat or /v1/messages call on its whole history, not from a compaction-typed message
* fix(proxy): trim a failed call's input at the newest compaction only when the client called /v1/responses
* fix(proxy): trim a failed call's input at the newest compaction when the proxy dispatched it as a Responses call
* test(proxy): run the dispatched-call-type failure test through the real request setup
* fix(proxy): count a failed Cursor stream's input from the newest compaction item
The Cursor route maps to no call type, so the stream generator's failure hook fell
back to the route and counted the whole history. Pass the processor's route_type.
* fix(proxy): count a broken Cursor stream's prompt from the newest compaction item
* fix(responses): keep a tool call a later output answers when trimming before a compaction item
* fix(responses): move a compaction-split tool call's turn to right before its output on both compaction routes
* perf(responses): look up moved turn items in a set so a long compaction-split turn trims in linear time
* test(responses): measure the compaction-split turn move as a CodSpeed benchmark instead of a clock read in unit tests
* fix(responses): move only calls still open at the compaction, and bill a disconnected stream from the newest compaction item
* fix(responses): restore only the newest stored call when a reused call_id is answered after a compaction item
* fix(spend logs): bound a Responses compaction item's summary like a chat block instead of cutting its encrypted_content
* test(proxy): price a failed dispatched call through a real ProxyLogging instead of mocked hooks
* fix(responses): finish a stream whose Claude compaction failed and group answered call ids by turn once
* fix(responses): count input_tokens after a compaction item the way inference sends it
* fix(llms): sanitize trailing slashes in provider get_complete_url
* test(llms): move trailing-slash url tests into each provider's mapped unit file
* fix(llms): strip trailing slashes in Perplexity search and Ollama embedding URLs and run the Topaz tests in CI
* fix(llms): strip trailing slashes in search, DashScope, VolcEngine, GigaChat and Gradient AI URL builders
* fix(fastcrw): strip a trailing slash from api_base and expect one slash in the GigaChat tests
* fix(gigachat): strip a trailing slash from the pass-through api_base
* fix(llms): join api_base and route with one shared helper
A deployment api_base that ends in / made several get_complete_url builders emit a double slash, which Envoy and other gateways answer with 404. Add join_api_base in url_utils and use it in those builders, with a regression test that drives every get_complete_url under litellm/llms with a trailing-slash api_base
* refactor(llms): move per-provider api_base strips onto join_api_base
Use the shared join_api_base helper in the search, chat, embedding, image variation, GigaChat and VolcEngine URL builders instead of a per-provider rstrip and suffix check, so one helper owns the single-slash join
* ci: run the trailing-slash URL builder test in the llms shard
* test(integration): cover trailing-slash api_base across provider URL builders
Wire cells for chat, streaming, messages, responses, embeddings, search and pass-through with and without a trailing slash, an owned-proxy module for the flag-gated providers, env bases and a worker kill, two chaos bursts, and a Topaz SDK cell
* fix(llms): send every provider route through join_api_base
Every place under litellm/llms and litellm/main.py that joined a caller-settable api_base with a route by concatenation or f-string now goes through join_api_base, so an api_base with a trailing slash sends the same upstream path as one without it
Routes made only of a caller-supplied id, model, or passthrough path stay on a plain join over an already normalized base, because the helper's suffix dedup would turn a lookup by id into a list or create call
The sweep test now also reaches the factory-made JSON provider builders, and a new loopback test drives the public entrypoints with an api_base ending in /gw/ and checks that every request path keeps /gw/ with no double slash
* fix(llms): keep the route skip only where a builder already had it
join_api_base is a pure join again. The new ensure_api_route skips the route only when the URL path
already ends in it, and it is used only where the merge base or the trailing-slash test PR head
already skipped. Evals cancel, Predibase, and Azure AI threads append their route again, EdenAI
model listing and Gemini RAG ingestion join through join_api_base, and Azure AI Anthropic checks the
path so a host named anthropic-* builds /anthropic/v1/messages
* fix(azure_ai): read /anthropic from the URL path when building the count_tokens URL
* test(llms): run the Gemini RAG trailing-slash case through aingest
The sync ingest wrapper needs a current event loop, which an earlier async test in the same worker clears
* test(llms): cover the Azure realtime, Bedrock rerank, and oobabooga joins with a trailing-slash api_base
* test(integration): cover api_base joins across providers, query strings and named hosts
* feat(url_utils): add join_api_base and ensure_api_route helpers
* fix(url_utils): keep the scheme slashes when a stripped suffix was the whole host
join_api_base and ensure_api_route now keep the // of a base left with only a scheme,
so http://v1 with /v1 stripped joins to http://v1/files as before. The Azure v1 client
base_url goes through ensure_api_route so a base ending in /openai/v1 no longer doubles it
* fix(azure_ai): build the count_tokens URL from the messages URL
* fix(azure_ai): match anthropic as a whole path segment in the messages URL
* fix(llms): keep main's Azure AI anthropic match and non-string Bedrock endpoints
Azure AI chat and messages go back to cutting the base at the first
/anthropic, as on main, and count_tokens keeps its path check, so a
gateway path like /anthropic-proxy builds the same URLs as main. A base
that already ends in /v1/messages still gets count_tokens next to it.
A non-string aws_bedrock_runtime_endpoint (an httpx.URL VPC endpoint) is
used as its text again in AgentCore, Bedrock responses and the RAGFlow
base, instead of falling back to the default host.
get_volcengine_base_url and GigaChat's get_api_base strip a trailing
slash again for outside callers. The cache-hit wire test is skipped as a
known proxy bug, since one cache hit writes two cache_hit spend rows.
* fix(llms): drop a slash before the api_base query and tighten the route test checks
* test(llms): annotate new test locals as Final and flatten the builder generator
* fix(url_utils): keep a scheme-only api_base hostless so it fails before sending
---------
Co-authored-by: Raushankumar0720 <raushan.singh.cg@gmail.com>
A guardrail hook plants a litellm_metadata key on a chat completions request after the proxy
stamped litellm_received_at into metadata. get_litellm_metadata_from_kwargs prefers that key
when both exist and only copied the user_api_key_* fields across, so the response timing
headers lost their receive anchor and x-litellm-overhead-duration-ms reported 2 ms on
/v1/chat/completions against 58 ms on /v1/responses for the same proxy work. The receive stamp
now rides along with the user_api_key_* fields.
Co-authored-by: devin-ai-integration[bot] <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* test(perf): add small_bodies_30rps and mixed_size_requests_4rps load cases
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* test(perf): run the load cases at 50 rps small and 5 rps mixed, the pod's knee
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* test(perf): ramp the load cases until the pod stops keeping up and report sustained_rps
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* test(perf): hand the ramp to the observation and widen the mixed ramp to 14 rps
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* test(perf): floor the small ramp at 20 rps and emit the per-step readings
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* test(perf): emit the ramp's hold bound and each step's unexpected statuses
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* test(perf): read each ramp step's cpu, redis latency and connections from the samplers' windows
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* test(perf): floor mixed_size_requests_ramp at 8 rps from the lane's 10 rps sustained step
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* test(perf): compare step readings with math.isclose so the e2e basedpyright gate stays clean
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
---------
Co-authored-by: kerry <kerry@berri.ai>
Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(proxy): keep a retried /v1/messages stream's spend log by logging the superseded attempt as a failure
A /v1/messages stream the router retries before any content shares one logging object across both attempts. The superseded attempt's end-of-stream logging ran as a success (a usage-only or placeholder result) and claimed the request's one-shot success log, so the recovered stream's spend row was either missing or written under the superseded attempt's message id. The passthrough logging path now logs an Anthropic stream whose frames carry an `event: error` as a failure with whatever usage the provider reported, the way a dropped stream already does, and the router closes the superseded stream before it opens the retry so that logging is queued first.
* fix(proxy): log a provider error frame through an attempt-scoped logging copy
The error-frame failure stashed the superseded attempt's usage on the logging
object the retry shares, so the retry's success log carried the failed
attempt's token counts. The failure now logs through a copy of the logging
object with its own model_call_details, and the helper moves into the
Anthropic passthrough logging handler so the proxy streaming handler only
routes by endpoint type.
* fix(proxy): scope the error-frame failure log to streams closed at the frame
A stream that ran to its end at the provider's error frame kept logging
through the success route before this PR, and the attempt-scoped failure
log is now taken only for a stream the router closed at that frame, with
the copy made at the close so the retry's deployment never reaches it
* fix(router): log each dropped /v1/messages attempt through its own logging copy and open the retry without its usage
A /v1/messages stream dropped after message_start and retried in its model group shares one Logging object
across attempts, so the retry's success log inherited the dropped attempt's stashed usage and exception,
and a second drop never reached the failure handlers because the first had already marked the object as
logged. The dropped attempt now dispatches its failure handlers through an attempt-scoped copy, and the
router discards the stashed usage right before it opens the next attempt, on same-group retries and on
every fallback hop, so a request that runs out of attempts still bills the last attempt's partial usage.
* fix(router): dispatch chat stream failures through the attempt-scoped logging copy
The router now clears a dropped attempt's stashed usage before every fallback hop, so the chat stream wrapper's queued failure log must read that usage from its own copy of the logging object, like the /v1/messages path already does. The stash test asserts through a real success log, the cooldown test reads a real Router's cooldown list, and the integration helpers carry full types.
* test(vertex): give the stream wrapper's logging stub an attempt_scoped_copy
* test: route the pass-through logging mock's attempt copy to itself and read the callback sink through a reader
The pass-through failure test stubbed dispatch_failure_handlers on a spec'd MagicMock, which the attempt-scoped copy no longer reached, so the logging worker never finished. The integration tests kept earlier sink batches in a caller-owned list the helper mutated; a small reader keeps them instead.
* test(router): use real Logging in fallback cooldown test
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(router): preserve superseded retry usage for billing
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(proxy): bill retry input after prior stream usage
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(proxy): track retry dispatch for terminal billing
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* test(router): use deterministic retry timestamp
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* test(router): avoid real clock in fallback tests
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(tests): remove duplicate streaming logger import
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(proxy): retain error-frame usage for exhausted retries
* test(proxy): fix retry billing expectations and helper annotations
* fix(proxy): preserve failure estimate without carried usage
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* refactor(proxy): audit retry helpers and fixtures
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* test(proxy): fix retry fixture timestamps
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* test(proxy): observe retry stash at attempt boundary
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(logging): keep terminal failure attribution when dropped attempts log on copies
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(logging): publish a dropped attempt's payload only while it is still the current attempt
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* test(proxy): cover spend-log encoding error propagation
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(anthropic): skip adapter failure dispatch when the wrapped stream already logged it
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(router): cool down the final fallback attempt when its failure callbacks were blocked
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* test(anthropic): import Final in adapter regression
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* test(proxy): drop spend-log encoding expectation superseded by main's poison isolation
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(anthropic): finish the wrapped stream's failure log before surfacing the error
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(anthropic): wait for the wrapped stream's failure log to finish like the inline dispatch did
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* test(anthropic): wait for the executor-run sync failure callback separately from the async caller boundary
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* test(proxy): pin pass-through stream fixture start times
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(logging): publish failed attempt payload before awaiting failure callbacks
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(logging): publish failed attempt payload from the failure handler after baseline invalidation
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* refactor(logging): drop unused discard_partial_usage_for_failure
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>
* feat(vertex-ai): support gemini 3.1 flash tts
* fix(vertex-ai): keep TTS metadata compatible with the cost map guard
* fix(vertex-ai): preserve existing Gemini TTS models and voice mappings
* fix(vertex-ai): recognize provider-stripped TTS model names
* test(vertex-ai): match current speech HTTP error response
* fix(vertex-ai): preserve nested speech locale and voice
* fix(vertex-ai): use provider speech media type in proxy
* fix(gemini): preserve TTS dispatch and nested speech settings
* test(gemini): run TTS regressions in CI shards
* fix(vertex-ai): account for Gemini Cloud TTS audio cost
* fix(vertex-ai): bill Gemini TTS without optional audio decoder
* test(gemini): cover TTS mapping and billing edge cases
* test(vertex-ai): cover Gemini voice normalization in provider shard
* fix(vertex-ai): keep Gemini TTS routing with generic provider metadata
* chore(vertex-ai): remove obsolete lint suppressions
* fix(gemini): strip audio format from speech configuration
* chore(vertex-ai): remove obsolete mutation suppressions
* fix(vertex-ai): preserve FLAC billing and wrapped audio headers
* fix(vertex-ai): reject unmeasurable FLAC before billing
* fix(vertex-ai): reconcile speech support with current provider mappings
* fix(vertex-ai): validate speech mappings and usage types
* fix(vertex-ai): preserve speech mapper override forwarding
* fix(gemini): map voice aliases across speech request forms
* fix(vertex-ai): send only formats the speech bridge cannot serve to Cloud TTS
Gemini TTS detection now goes by model name, so newer Vertex Gemini TTS
models keep working, and only mp3, opus, alaw, mulaw and other container
formats move to Cloud Text-to-Speech. wav and pcm requests stay on
generateContent as on main.
A Cloud TTS response whose duration cannot be measured now returns the
audio unbilled with a warning instead of raising after a paid call. FLAC
handling is removed because Cloud TTS has no FLAC encoding.
The cost map, the google_genai mappers and the proxy error path go back
to main, since those changes were outside this feature.
* fix(vertex_ai): keep Gemini TTS formats, region and instructions on routes that serve them
Only mp3, opus, alaw and mulaw go to Cloud Text-to-Speech. Every other
format gets a local 400 that lists the supported ones, before Google is
called.
Cloud TTS uses the deployment's us or eu regional host. Other regions
and a custom api_base get a 400 that says to request wav or pcm.
instructions is sent as the Cloud TTS style prompt, and a request with
no voice keeps the Gemini model.
get_usage_object is back to main's code, so dict usages still go
through Usage.__init__, and the proxy no longer labels pcm16 or
linear16 output as WAV.
* fix(vertex_ai): send Gemini TTS to every regional Cloud TTS host and keep speak tags as text
Cloud Text-to-Speech serves Gemini TTS on a host named after each Vertex
location, so a non-global vertex_location now uses
{location}-texttospeech.googleapis.com instead of only the us and eu hosts.
Google answers for regions it does not serve.
Gemini TTS voices reject SSML, so input that contains <speak> is sent as
text for Gemini TTS models. Chirp keeps the SSML auto-detect.
Removes the proxy alaw and mulaw media-type entries. Cloud TTS returns
those encodings with a WAV header, so the response is labeled audio/wav
and the entries never applied.
Adds a regression test that custom per-token pricing bills the usage a
speech response reports.
* fix(vertex_ai): map Gemini TTS regions to Google's Cloud TTS locations and keep a bad voice from cooling down the deployment
US and EU Vertex regions go to the us and eu Cloud TTS hosts, northamerica-northeast1 to its own, global or unset to global, and any other region gets a local 400. A Cloud TTS 404 on the Gemini path, such as a misspelled voice, comes back as a 400 so the router keeps the deployment serving wav. Usage reads the sent request from the response instead of the logging object.
* fix(vertex_ai): default Gemini TTS Cloud formats to a Gemini voice and validate speakers
Parse only usage fields from the sent Cloud TTS request, so a numeric speakingRate no longer fails after Google bills it.
Send the Gemini voice Kore when no voice is named, and reject Cloud TTS voice names unless drop_params is set.
Reject malformed speaker entries unless drop_params is set, and translate Cloud TTS speaker fields on wav and chat audio.
Use the Vertex default region when no location is set, so mp3 and wav share a region.
* test(vertex_ai): cover two-speaker Gemini TTS mp3 through aspeech with fake credentials
Replaces two tests that patched the Vertex TTS config's own token and URL methods.
* fix(vertex_ai): reject malformed Gemini TTS voices and keep beta Chirp and Lyria params unmapped
* fix(vertex_ai): drop unknown Gemini TTS voice keys on wav and pcm instead of rejecting them
* fix(vertex_ai): reject unknown Gemini TTS voice keys on wav, pcm, and chat audio unless drop_params is set
* fix(vertex_ai): honor every drop_params setting in the Gemini TTS voice key check
Global drop_params, batch row and deployment drop_params on Vertex batch
file uploads now drop unknown voice keys. A voice object with both
speechConfig and speech_config gets a 400 for the ignored speech_config.
* fix(gemini): drop unknown nested voice fields under drop_params and detect TTS by served model
* fix(gemini): filter unknown speech config fields with fixed-depth helpers and pass audio params through unchanged
* fix(gemini): camel-case snake replicated voice keys so drop_params keeps them
* fix(vertex-ai): reject voice fields Cloud TTS cannot send unless drop_params is set
* fix(vertex-ai): reject wrong-typed Cloud TTS voice values and check speakers as sent
Gemini TTS on mp3, opus, alaw, or mulaw now answers 400 for a voice field whose value Cloud TTS cannot send, such as a list languageCode or a string voiceConfig, and for nested voice fields under a speaker that also names speakerId. drop_params still drops them.
* fix(vertex-ai): reject wrong-typed speech config wrappers and speakerId overriding a prebuilt voice on Cloud TTS
* fix(vertex-ai): keep speakers and voice_config beside a speech_config wrapper on Cloud TTS
* fix(vertex-ai): pick Cloud TTS voices from the same speech config the WAV path sends to Gemini
* fix(files): carry the deployment's drop_params on model= file uploads so Vertex batch rows drop unknown voice keys
* fix(gemini): keep a structured speech voice when the model is a model_alias_map alias for a Gemini TTS model
---------
Co-authored-by: mateo-berri <277851410+mateo-berri@users.noreply.github.com>
* fix(cli): send pi's session id from lite pi so sessions group and stick
* docs(cli): drop the provider_block docstring lines the README already covers
* docs(cli): note the session id format the proxy reads for lite pi
* test(integration): keep autovacuum off the size cleanup test's scratch tables
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* ci(circleci): keep the MCP SDK1 peer venv on mcp 1.x despite the project override
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>
* test(responses): cover billing every round of a streamed MCP auto-execute
* fix(responses): bill every round of a streamed MCP auto-execute on one spend row
* fix(responses): bill every round of a streamed MCP auto-execute on bridged providers
A streamed /v1/responses MCP auto-execute on a provider bridged through chat
completions (for example Anthropic) billed only round 1, because each round's
chat stream dispatched on the shared logging object and the first dispatch won.
Bridged rounds now go through the same one-row billing as native rounds. A
round is billed as soon as its chat stream ends, before its closing events
reach the client. The row carries the final response, usage and cost summed
over every round, and the fixed margin once. A round served from the response
cache is billed at zero. A follow-up chat stream that breaks bills the paid
rounds on its failure row. A single-round stream logs exactly as before.
* fix(responses): charge one fixed margin and keep bridged usage fields on streamed MCP rounds
A follow-up round that fails after earlier rounds were billed now charges the
per-request fixed margin once, and its failure row carries no cost breakdown
that would disagree with the summed cost.
A bridged multi-round stream logs a plain cache_hit flag like a single-round
stream does, and its usage keeps the summed provider usage fields of every
round.
A new test covers a stream dropped after its event loop closed, which must
release its parked log without an unawaited coroutine warning.
* fix(responses): price each streamed MCP round on its own and bill every round when a guardrail blocks
Each round of a streamed MCP auto-execute is now priced from its own
uncached cost, so a provider that states its cost (xAI) charges the stated
cost of every round plus the fixed margin once, and the cost breakdown
always agrees with the billed cost.
Summed usage goes through a JSON-safe dump, so a nested provider field such
as xAI's context_details no longer breaks a failed stream's response.failed
event.
A post-call guardrail that blocks a multi-round bridged stream now drops the
parked success log and bills every round on the failure row, and a failed
follow-up round's failure row carries the summed breakdown of both rounds.
* fix(responses): keep earlier MCP rounds on their success row when a later round fails
When a later round of a streamed MCP auto-execute fails, the rounds before
it are now logged on their success row, and the failure row bills only the
failed round with no second fixed margin. That failure log runs on its own
copy of the request's logging state, so its usage and cost never leak into
the earlier rounds' row.
A post-call guardrail that blocks a native multi-round stream keeps the
summed success row beside the $0 failure row, as at the merge base, while a
bridged stream still bills every round on its failure row.
* fix(responses): keep a failed later MCP round on its own failure row
A streamed MCP auto-execute whose later round failed wrote only the earlier
rounds' success row. The failure was logged on a copy of the logging state,
while the proxy builds its failure row from the request's own state, which by
then held the success payload, so the failure row took the success row's id
and was dropped.
Now the failure is logged on the request's own state as at the merge base,
and once a later round starts, the success row for the rounds before it is
logged from a copy of the state taken at that moment, so the two rows never
mix.
* fix(proxy): let an MCP tool call reuse its request's max_parallel_requests slot
* test(proxy): create the other-request task in a way Python 3.10 supports and mark its bindings Final
* fix(proxy): keep a reused max_parallel slot until the request's last round
An MCP tool call that reuses its request's max_parallel_requests slot no
longer loses it to the first round's late success callback, so the key
stays counted for the rounds still to run. The reused call also stops
replacing the request's rate limit headers.
* fix(responses): keep an unpriced streamed MCP round unpriced instead of billing it as free
* fix(responses): log a bridged MCP stream as a chat completion with every round's usage
* feat(ui): add Delete to the guardrail detail Settings card for DB guardrails
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* test(ui): hoist guardrail delete test fixtures to stay within the inline-object lint budget
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* feat(ui): share one guardrail delete modal between the table row and the detail page
The table-row delete now has the same dangling-reference warning and typed-name
confirmation as the detail page. Tests now assert the success and error toasts
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(ui): warn that deleting a guardrail used in a policy pipeline step blocks requests with a 400
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* test(ui): cover hiding guardrail Delete while settings are being edited
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>
* feat(gemini): LITELLM_GEMINI_DEFAULT_MAX_OUTPUT_TOKENS sets the long decode budget for requests that send no limit
* test(gemini): cover a default long decode budget set by LITELLM_GEMINI_DEFAULT_MAX_OUTPUT_TOKENS
* fix(gemini): apply the default output budget only to requests asking for text output
* refactor(gemini): apply the default output budget once in the async request transform
* fix(gemini): leave the default output budget off non-chat models and a snake_case generation_config
* fix(gemini): skip the default output budget for audio-only models and continuation resumes
Lyria and other Gemini chat models whose cost map row lists no text output answer 400 when sent a maxOutputTokens, and a request that resumes from a continuation token answers 400 when a limit is added beside it. Both now go out without the default budget.
* fix(vertex_ai): skip the Gemini default budget for a model a deployment's model_info overrides, and pass odd generationConfig shapes through
* fix(vertex_ai): cap the Gemini default budget per deployment at the loaded cost map and its own model_info limit
* test(vertex_ai): wrap a long test signature under the 120 character limit
* fix(vertex_ai): keep a direct call's default output budget under the loaded cost map limit
A litellm.Router deployment's model_info writes onto the shared cost map row, so a later direct call with no model_info read that raised value and Vertex answered 400. The direct call now caps at the lower of the loaded row and the live row, so litellm.register_model can still lower the cap.
* fix(vertex_ai): ignore a deployment output limit that is not positive
A model_info.max_output_tokens of 0 or -1 capped the default budget at that value, and Google answers 400 to a limit that is not positive. It now counts as unreadable, so the cost map limit applies.
* fix(gemini): never continue a request asking for more than one candidate in camelCase
* fix(gemini): keep continuing a request whose candidate count is null or 0
* fix(router): unregister strategy selectors from global callbacks on discard
Router.discard only removed its own bound methods from litellm's callback
lists, so the strategy selectors it registered (least-busy, usage-based,
latency-based, cost-based and routing group selectors) stayed behind and
kept receiving events after the router was gone. Reuse the identity based
unregister that update_settings already uses so a discarded router takes
only its own selectors with it
* fix(router): purge discarded selectors from every event callback list
function_setup copies litellm.callbacks into the success and failure
lists on every request, so a router that served traffic still leaked its
selectors there after discard. Route the cleanup through
remove_callback_from_all_lists, which now also covers input_callback, and
fold the selector gathering into discard so the router coverage check
sees no untested helper
* fix(router): keep live routers, in-flight hooks and user_config limits intact on discard
discard() now rebuilds every global callback list without the router's selectors, optional pre-call checks and own usage callbacks instead of removing them in place, so a callback loop already running never skips a hook. It then re-registers what the remaining live routers need, so a same-strategy survivor keeps its deduped handler. The proxy's per-request user_config router is discarded only after its call returns.
* fix(router): bind user_config handlers to the request and tighten discard
When a user_config request carries a logging object, its router is
discarded before the request runs and its handlers ride on that
logging object, so v1 rpm and v2 tpm limits count every request and
concurrent requests count once. Without one, the router is discarded
after the request as before. The router's pre-call loops also read
the request-scoped handlers.
discard() now refills only the handler keys it vacated, guards the
callback lists and the live-router set with a lock, covers routers
built without a model_list, and drops its Slack alerting, adaptive
router hook and budget limiter sync task
* fix(proxy): reuse one user_config router per config and bind its handlers to each request
Every user_config request now goes through a cached router built once per
config. Its strategy handlers leave the global callback lists and are bound
to the request's logging object instead, so limits count each request once
and nothing is left behind after the request.
* fix(router): keep the proxy budget limiter on user_config requests
A user_config router now swaps in only its strategy selectors, so the proxy's
RouterBudgetLimiting stays on the request and keeps counting its spend. Usage
logged after a strategy is retired, such as a request still in flight when its
user_config router is evicted, now schedules one more push to Redis.
* fix(proxy): init the retire fields on the key model budget limiter
The key model budget limiter builds its own state instead of calling the base
init, so with the proxy's Redis attached every spend increment raised on the
missing retire flag and the key's model budget never counted spend.
* fix(router): bind the deployment callbacks to user_config requests
A cached user_config router serves requests through each request's logging
object but bound only its selectors and policies there, so a failing
deployment was never cooled down and its usage counters missed requests.
* fix(router): retry an evicted router's budget writes and keep the adaptive header on user_config requests
A retired budget limiter now keeps retrying its queued Redis increments every
sync interval until Redis accepts them, instead of giving up after one failed
write.
The x-litellm-adaptive-router-model header now comes from one stateless hook in
litellm.callbacks that holds no router, so user_config adaptive routers, whose
per-router hooks are request-scoped, still answer with the header.
* fix(router): lock request callback setup against router cleanup and drain in-flight budget writes on retire
function_setup now edits the global callback lists under the same lock that
router cleanup uses, so a router dropping its callbacks on another thread can
no longer shift the indexes function_setup pops or put a dropped handler back.
A budget limiter retired while a Redis write is still in flight now keeps
retrying that batch if the write fails, instead of leaving the spend queued
with no task left to send it.
* fix(utils): set up loggers after releasing the callback lists lock
function_setup held the callback lists lock while set_callbacks ran its one-time
logger setup, which can block on a Slack package install, so other threads and
Router.discard waited on it. The lock now covers only the list reads and writes.
* test(router): annotate the retire test's locals as Final
* fix(logging): skip replaced global callbacks on the input path too
replace_global_callback already kept a same-key global logger out of
success and failure dispatch. pre_call and post_call still sent it the
input events, so a logger with an input hook could count a request start
it never saw finish. Both now share one filter with the success path
* test(logging): type the input-path regression test
---------
Co-authored-by: mateo-berri <277851410+mateo-berri@users.noreply.github.com>
* fix: Send the include_usage chunk with an empty choices array
Signed-off-by: Alberto Perdomo <aperdomo@redhat.com>
* test: Skip chunks with no choices when reading streams
Signed-off-by: Alberto Perdomo <aperdomo@redhat.com>
* fix: Strip usage from plain-iterator async stream chunks
Signed-off-by: Alberto Perdomo <aperdomo@redhat.com>
* test: Expect empty choices on the usage chunk in the gemma tracing test
Signed-off-by: Alberto Perdomo <aperdomo@redhat.com>
* fix(streaming): send the empty-choices usage chunk only to callers that asked for include_usage
The usage chunk now has choices=[] only when the caller set
stream_options.include_usage itself. When the OpenAI handler, the proxy's
always_include_stream_usage setting, or the harness added include_usage on the
caller's behalf, the chunk keeps its legacy single empty-delta choice, so
clients that read choices[0] on every chunk keep working.
The proxy and harness mark their injection with the private
_litellm_stream_usage_injected flag, and the proxy scrubs a client-sent copy.
Plain-iterator streams keep usage on their chunk when usage was not requested,
and the LLM Shield trailing flush ignores chunks with no choices.
* fix(proxy): mark the default stream usage injection so hooks keep the legacy usage chunk shape
The proxy's default strip-mode injection adds include_usage for clients that did not ask and drops the usage chunk before it reaches them, but streaming iterator hooks still saw it with choices=[]. Marking that injection with _litellm_stream_usage_injected keeps the old single empty choice for them.
* fix(router): treat deployment and router-default stream_options as injected, and drop the client flag on the queue route
* refactor(router): move the stream usage injection check into router_utils
* fix(router): read stream_options from pydantic deployment params when marking injected usage
Update the mantle, openinference, and text-completion integration tests to expect the include_usage chunk with empty choices.
* fix(proxy): keep the stream usage injection marker out of stored request bodies
* test(streaming): cover who asked for stream usage on chat, queue, deployment and spend log paths
* test: type the queue flag test, pin the logging start time, and build the deployment usage config without mutation
---------
Signed-off-by: Alberto Perdomo <aperdomo@redhat.com>
Co-authored-by: mateo-berri <277851410+mateo-berri@users.noreply.github.com>
Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
The OpenAI fine-tuning handler built its SDK client from every local it
received, so an api_version or litellm_params set on an openai
finetune_settings entry reached OpenAI.__init__ and every fine-tuning
call on that provider failed with "unexpected keyword argument
'api_version'". Build the client from its named params instead and
cover it with a request-level regression test
Co-authored-by: mateo-berri <277851410+mateo-berri@users.noreply.github.com>
* fix(token_counter): stop re-downloading a HuggingFace tokenizer after a failed download
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* test(tokenizer): isolate HuggingFace tokenizer caches
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(token_counter): clear the traceback before re-raising a cached tokenizer load error
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(token_counter): retry failed HuggingFace tokenizer loads after the cache window
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(token_counter): start tokenizer retry TTL after a failure
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(utils): lock every HuggingFace tokenizer load and keep load failures as values
Concurrent llama-2, llama-3 and cohere counts each started their own failed
download because only the bundled tokenizers took a load lock. Every tokenizer
kind now has one, so a burst during an outage makes a single attempt and two
counts can no longer read the same expiring entry at once.
A failed load is now remembered as a small value with the error message
instead of the error object, so re-raising it no longer keeps the first
failing request's frames alive. Adds tests for the concurrent outage and for
fork-guard errors staying raised and unremembered.
* fix(utils): keep Hugging Face tokenizer downloads unlocked and warn once per failed download
Reverts the per-kind load lock so concurrent first downloads still run in parallel, as #47353 intends. A short lock around the failure cache read and write stops two counts racing on an expiring entry, and the tiktoken fallback warning now fires only when a download fails, not on every count that reuses the remembered failure.
* fix(caching): read an entry another reader just expired as a cache miss
Two threads reading the same expiring key in InMemoryCache could both pass the membership check, and the second then raised KeyError after the first removed it. get_cache now treats that key as a miss. This replaces the extra lock the tokenizer failure cache took, and the regression test forces the interleaving without timing.
* fix(caching): treat a TTL another reader just removed as not expired
InMemoryCache's expiry check read the TTL after a separate membership check, so a reader racing another reader's eviction raised KeyError. It now reads the TTL once and treats a missing one as not expired, and get_cache already reads the vanished entry as a miss.
---------
Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Co-authored-by: mateo-berri <277851410+mateo-berri@users.noreply.github.com>
* fix(scim): write a group's roster in one locked transaction instead of one member at a time
POST, PUT and PATCH /scim/v2/Groups resolve every member in one query and write the roster, the
membership rows and the users' team tags in one transaction under the team's advisory lock, so a
500-member push runs a fixed number of statements instead of thousands. PUT now answers with the
roster it just wrote
* test(scim): assert the roster the IdP gets back and boot the bulk roster cell with two workers
* fix(scim): skip the team lock for a rename-only PATCH and lock member rows in id order before the roster write
A PATCH naming no members opened the roster transaction and took the team advisory lock for nothing. A SIGKILLed worker then left its orphaned query engine holding that lock until the 60 s transaction timeout, and /team/delete, which takes the same lock, timed out behind it. An empty delta is now answered from a plain team read, with no transaction and no lock
Two groups trading members in parallel locked user rows in opposite orders, and one request died with a Postgres deadlock. Every user row a push touches is now locked with SELECT ... ORDER BY user_id FOR UPDATE before the first write, so the two requests serialize
The bulk roster cell gets a 120 s client through a new owned_proxy_process client_timeout, so an unfixed push fails on the statement count instead of the harness clock, requires positive counts on both pushes, and gains a parallel roster-trade test that answered 500 before the lock
* fix(scim): evict the cached team on a group rename and read a no-op roster from the writer
A rename-only PUT or PATCH wrote the new alias and left the team cached under
its id and old alias on every worker, so keys kept logging spend against the
old team alias until the TTL expired. The group write now evicts the cached
team (and broadcasts the eviction) right after the alias or metadata update.
sync_team_roster reads the team from the writer when the plan names nobody,
so a replica cannot hand back the pre-rename row; a plan already matching the
roster returns inside the lock without writing, evicting, or emitting the
members metric; the response is built from the written row, so lastModified
reflects the roster write; and detaching a removed member bumps the user's
updated_at like every other user write.
* fix(scim): bump detached members' updated_at in UTC
The detach UPDATE wrote updated_at = CURRENT_TIMESTAMP, which Postgres casts
into Prisma's naive timestamp column in the session timezone, so on a database
whose session is not UTC the removed member's row landed hours away from the
UTC instants Prisma itself writes. Write NOW() AT TIME ZONE 'UTC' instead, and
have the bulk roster cell assert the bumped row sits between two reads of the
database's own UTC clock taken around the replace.
* fix(scim): delete a group through the team delete's bulk path instead of one member at a time
* fix(scim): chunk the group write's member lookups under the bind cap
Every Prisma "in" filter the group write issues bound one parameter per
member with nothing bounding the list, which check_unbounded_in_lists
reports as 11 new findings. The single-field lookups now go through
litellm.repositories.chunked_in (caller_transaction inside the roster
transaction, per_chunk_ok for the role recompute), and the one
cross-field OR read slices its values at IN_LIST_CHUNK_SIZE and records
the bound. A push of IN_LIST_CHUNK_SIZE + 1 members now reads in two
slices, and a member named only in the last slice still resolves.
* fix(scim): demote a deleted admin group's members before the team delete so a failed delete stays retryable
* fix(scim): recompute member roles again after the team delete so a concurrent group write cannot regrant admin
* fix(proxy): send bulk cache evictions in slices so the broadcaster drops none
Veria pointed out that the invalidation publisher caps its pending tasks at
1,024 and drops the excess with a warning, and that a group push now detaches
every removed member's keys in one call instead of one member at a time, so a
removal of more than roughly 340 members (three cache keys each) silently left
stale key objects on the other replicas. The merge base never hit the cap
because each member was a separate request.
Bulk evictions (team, user, organization, and session key deletes, plus the
roster settle) now publish in slices of 256 and wait, bounded at two seconds,
for the backlog already handed to the event loop before each slice. A Redis
that never answers costs a bulk caller at most that wait per slice instead of
wedging the request, and single-key callers are untouched. Regression tests
push 1,500 keys and a 500-member removal through a slow publisher and assert
every eviction reaches Redis; both fail against the previous tip.
* fix(proxy): clear this worker's cache before pacing bulk eviction broadcasts
Greptile pointed out that pacing the broadcasts in slices also paced the
local eviction: a bulk delete waited for the publish backlog before dropping
the next slice from this worker's own memory, so on a Redis that accepts
connections but never answers a key deleted from the database kept
authenticating on the handling worker for up to two seconds per slice. The
merge base cleared every key locally at once.
Bulk evictions now drop every affected entry from this worker's memory first,
through the cache's own partition lookup since key objects live apart from
the other auth objects, and only then publish in paced slices. The roster
settle does the same for the deleted tokens, the JWT mapping keys and every
changed member's user and membership entries before any step that can wait.
Regression tests run both paths against a publisher that never answers and
check that nothing is still cached on this worker before the first bound
elapses; both fail with the local eviction removed.
* fix(scim): write recomputed member roles only where the membership still backs them
* fix(team): report the users a team delete detaches so SCIM demotes late-added members
* test(scim): audit cells for the bulk group delete
Move the SCIM group helpers into tests/integration/_support/scim.py and add the delete cells: a member's other
group survives, the admin group's members are demoted and the audit row names the SCIM caller, a member a
concurrent PUT adds while the delete waits on the team lock is demoted too, unknown, repeated, non-admin and
hostile ids, an empty group, the peer proxy, a 20-delete burst across both proxies, and a delete after a worker
was SIGKILLed
* fix(proxy): drop the Redis copies of bulk-evicted auth objects before pacing the broadcasts
* docs(proxy): drop the module docstring that repeated the PR description
* fix(team): sweep deleted team references only under the lock so a late-added member is demoted
The unlocked sweep that ran before the team's advisory lock stripped a member a
concurrent group write added after the roster was read, so neither the roster
snapshot nor the locked sweep's RETURNING named them and the SCIM recompute
left them proxy_admin after the admin group was gone. The locked sweep does the
same strip and reports whom it stripped, so it is the only sweep now.
The integration cell holds the delete on a member's user row so a PATCH lands
between the roster read and the lock, then checks the late member is demoted.
Both waiter queries in the SCIM support module return their count as text so
the first poll parses.
* chore(team): carry the unbounded-IN baseline entries to delete_validated_teams
The delete pass moved from delete_team and the unlocked sweep into delete_validated_teams, so the checker's baseline names the function the two team_id IN lists now live in
* test(scim): expect the master key alias in the audit row and send post-kill requests on a fresh connection
* fix(scim): bound bulk cache evictions to one Redis round trip and close lock-order gaps
Bulk key evictions now drop entries from memory, delete them from Redis in one
chunked DEL, and broadcast every key in one pipelined publish, so a large roster
removal no longer overflows the publisher's pending cap or costs a Redis round
trip per member. A push waits for in-flight membership loads before evicting.
User rows are locked in byte order to match the spend writer, a bulk user delete
takes every team lock before its first user write, and a POST whose roster write
fails deletes the empty team so the IdP retry creates it again.
* test(team): observe delete broadcasts at the Redis layer
* fix(scim): lock user rows in id order on every multi-row writer and create unprovisioned members concurrently
* fix(team): chunk the email lookup behind the member row lock
* fix(scim): create a member deleted mid-push again instead of failing the group write
* fix(scim): refuse strict-mode mid-push deletes with 400 and check license seats once per batch
A member deleted mid-push is created again only when scim_upsert_user allows it,
under the verb's own created_via. New group members are checked against the
license seat limit as one batch before any is created. Auth cache publish drops
log the key count instead of the key.
* fix(scim): lock member rows while recomputing admin roles after a group push
* fix(users): re-read teams under the row locks in bulk user delete so a racing group push keeps no deleted member
* fix(users): take /user/delete and SCIM DELETE /Users users off their current teams under the row locks so a racing group push keeps no deleted member
* fix(scim): keep retrying a push while deletes race it, lock rows in member bulk delete, evict removed keys from Redis first
A group push gave up after one retry when SCIM user deletes kept landing between attempts, answering 500 in 9 of 10 race rounds. It now retries while each attempt finds newly deleted members, and gives up only when a member it created again goes missing a second time.
The team member bulk delete now locks the removed members' user rows in the shared order and takes the team off their teams arrays in one array_remove statement, so it no longer writes back a teams array read before a concurrent push added another team.
A push that removes members now drops their deleted team keys from memory and Redis together with the members' other cache entries before the team eviction, so a removed key cannot refill from Redis mid-cleanup.
* fix(user-delete): keep the team alias label on the team members gauge after removals
* test(integration): count relay statements over plaintext so an SSL-enabled Postgres does not hide them
* fix(user-delete): skip a team deleted while the user delete waited on its lock instead of answering 404
* fix(scim): answer 404 and roll back when the team is deleted during a roster write
* test(integration): audit cells for SCIM bulk roster writes, user delete locking and the eviction broadcast
* test(scim): cover deleting a SCIM group through the team delete's bulk path
* fix(scim): evict a deleted group's member caches in bulk and drop cached roles after a recompute
* fix(scim): answer 404 to a group delete that lost a race to another delete
* fix(scim): chunk the locked team lookup and delete in the team delete transaction
* fix(scim): keep service-account keys and team models on a group delete and write its records once under the lock
* fix(scim): demote a deleted group's members inside the delete transaction
The admin-group role write now runs on the team delete's transaction, over the
roster read under the team lock, so a failed delete leaves roles untouched and
a member added while the delete waited is demoted with the rest. The delete
transaction also waits for a free connection as long as it may run, so a burst
of deletes queues instead of failing after Prisma's two-second default. Every
member in the 5-vs-500 statement test now holds a key.
* fix(audit): write deleted keys' audit rows in one insert
A delete that removes many keys wrote one audit insert task per key, and those tasks held pool connections the delete's follow-up writes waited on. The race test now waits on any lock wait behind the delete.
* fix(ui): show the budget spend limit and rate limits without a collapsed section
The create and edit budget dialogs hid max budget and reset period behind a collapsed Optional Settings section that sat 80px below the last field, and dropped both values from the request unless the section was open on submit
Both dialogs now share BudgetFormFields. Budget ID comes first, then the spend limit as one row ("$500 per month"), then the three rate limits as short fields beside their labels. A secondary Optional badge marks the spend limit and the rate limits group, so people can still tell those fields are optional. Every field is always submitted, and the edit dialog sends the existing spend limit and reset period back unchanged
* fix(ui): send only changed fields when editing a budget
Saving the edit budget dialog resent every field, so an unrelated rate
limit edit rounded a stored sub-cent spend limit to 2 decimals (0.001
became 0) and resent the reset period, which makes the proxy recompute
the next reset for custom durations. The dialog now sends the budget ID
plus the fields the user changed, and the proxy keeps everything else.
Also updates the budgets Playwright spec for the new field labels and
the removed Optional Settings section.
* fix(ui): track budget edit changes from render so a reverted field cannot hide a later one
react-hook-form's formState is a render-time snapshot. Reading dirtyFields
only inside the submit handler meant that after a field was changed and
changed back, a later edit to another field was missing from the first
save. Reading dirtyFields during render keeps the snapshot current.
* refactor(ui): rename the budget components to PascalCase
Renames budget_modal, edit_budget_modal and budget_panel (and their tests) to BudgetModal, EditBudgetModal and BudgetPanel, and drops their filename-pascal-case suppressions.
* feat(model hub): show admin-defined custom tags and metadata from model_info
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* refactor(router): move model hub custom field merging into router_utils
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* refactor(model hub): share the custom tag badge list between the column and details
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* test(model hub): cover public hub, patch, null, hostile, forwarding, and killed-worker paths for custom fields
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(model hub): drop custom_metadata entries with non-string keys or non-finite numbers
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* refactor(model hub): use isinstance narrowing instead of TypeGuard for custom_metadata values
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* refactor(model hub): build custom tags and metadata in the proxy instead of the router
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(model hub): serve custom tags and metadata on model_group_alias rows
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* remove docstring from model_group_custom_fields.py
* style(model hub): fix lint and format in custom metadata tests
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* style(model hub): drop leading blank line in custom fields helper
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>
* fix(ui): show a failed state with Retry in DataTable when the fetch errors instead of an empty table
* fix(ui): show the shared failed state with Retry on the Usage spend totals and the Caching activity chart when their fetch errors
* refactor(ui): name the settled-fetch objects so the inline object arg lint budget holds
* test(ui): find the caching chart card through the existing cardTitled helper
---------
Co-authored-by: devin-ai-integration[bot] <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(gemini)!: honor supports_sampling_params on the Gemini and Vertex path and flag Gemini 3.6 Flash, 3.5 Flash-Lite and later
Google ignores temperature, top_p and top_k on Gemini 3.6 Flash, Gemini 3.5
Flash-Lite and every later Gemini release, and says a later generation will
return 400 for them. The model map now carries supports_sampling_params: false
on those rows, and the Gemini and Vertex transformation reads the flag the way
the Anthropic and Bedrock paths already do, dropping the params under
drop_params and otherwise raising the client-side 400 that names drop_params
BREAKING CHANGE: a request that sends temperature, top_p or top_k to a flagged
Gemini model without drop_params now gets a 400 instead of a 200 whose
sampling params Google silently ignored
Fixes#47015
* fix(gemini): gate the passed-through top_k where the request body is built and keep allowed_openai_params working
top_k is a provider-specific kwarg that never reaches the OpenAI param mapping,
so the flag has to be applied while the Gemini request body is built, the same
boundary Anthropic uses. The mapping now skips a flagged sampling param instead
of raising, because the shared unsupported-param check already returns the 400
that names drop_params, and a param the caller lists in allowed_openai_params
is forwarded verbatim as that 400 promises
* fix(gemini): filter top_k out of the generation config instead of mutating optional_params
* test(gemini): type the sampling-flag twin test and keep its locals out of the loop body
* fix(vertex_ai): reject or drop sampling params on batch rows for flagged Gemini models
The batch JSONL wrapper calls map_openai_params directly and skips get_optional_params,
so a flagged model's temperature, top_p and top_k vanished from the uploaded row with no
400. Batch rows now get the same contract as a real-time call: a 400 naming drop_params,
or the knobs dropped when litellm.drop_params is on
* fix(vertex_ai): ignore null sampling knobs on batch rows for flagged Gemini models
A batch row carrying temperature: null, top_p: null or top_k: null means no
override, the same as on a real-time call, so the row is no longer refused
with a 400 and the null knobs are simply not forwarded
