The swe-1.7 rates were briefly lowered to the standard tier. The docs page
records the API-served swe-1.7 as the Cerebras-served Lightning tier, so put
the matching rates back rather than have the cost map and the docs disagree.
Cognition also answers /v1/responses through the chat-completions bridge, the
same as every other provider in the JSON registry, so the endpoints support
matrix should say so instead of under-declaring it.
The swe-1.7 rates were carried over from the closed prior attempt and
match SWE-1.7 Lightning, 5x the SWE-1.7 Max and Medium rates the vendor
publishes. swe-1.6 was already on the standard tier, so the two entries
disagreed with each other. Both now read 0.5 in, 2.5 out, 0.2 cached per
million tokens.
Also drops the redundant registry comment in constants.py.
The passthrough tests and their coverage registry rows pointed at the internal
ticket id, which does not resolve for anyone following a link from
status.litellm.ai. Each test docstring and registry rationale now names the
GitHub issue it pins: #36086 for the two prefix routing cases, #36087 for the
file list cursors, #36523 for streamed Responses cost, and #36646 for
embeddings spend.
Every pod schedules the budget reset job, so a fleet re-read the whole due
population and wrote it back against one Postgres at the same calendar
boundary, multiplying a single sweep by its replica count. The job now takes
the shared PodLockManager lease, so one pod sweeps per tick. A deployment with
no Redis keeps its previous behavior, and a Redis that cannot answer sweeps
unguarded rather than stranding every expired budget at its cap.
The per-window scan read every row carrying budget_limits in one statement, so
its cost grew with the deployment's key count. It is now keyset-paginated and
walks to the end of the table on every sweep. A per-run cap would need a resume
position, and no pod can hold one because the lease rotates between ticks, so
the strictly advancing cursor is what terminates the walk.
Found and updated rows were also JSON-serialized into the service hook's
metadata and into debug lines on every chunk, on the event loop, whether or not
anything consumed them. The hooks now carry counts, and the debug payload is
deferred until a record is actually emitted.
Resolves LIT-4793
Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
a369cb0da7 made the chat-to-responses bridge return the routing prefix on the
model it passes to responses(), so responses() re-resolving the provider is a
no-op instead of stripping a second prefix. It updated the bridge's own unit
tests but not this one, which still asserted the stripped id and has been
failing llm_translation_testing since that change landed.
The provider still receives gpt-5.4: responses() strips the openai/ prefix on
its own resolve, one layer later than this assertion used to sit. The stale
comment claiming the prefix is stripped before routing goes with it.
SCIM roster writes were swallowed, so a group or user push returned 200 while the
team roster never received the membership. Surfacing the failure fixes that, but
aborting on the first failed write leaves the rest of the batch unattempted on top
of unrolled-back, which is worse than what it replaces.
Every roster write in a reconciliation is now attempted, and the ones that did not
land are reported together, naming each failed add and remove. Rollback would be the
other option and it is not safe here: the compensating write can fail too, and it can
strip a membership that pre-dated the push. SCIM reconciliation is idempotent, so a
named partial failure is what the IdP's next push needs to close the gap.
The reported status still follows the failures, so a unanimous 404 stays a 404 and
only a batch whose failures disagree falls back to 500.
Co-authored-by: yassin <yassin@berri.ai>
Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
The read replica never received the operator's DB pool settings, so its
Prisma pool fell back to `num_physical_cpus * 2 + 1` and the configured cap
was not enforced. Both startup paths now pass the same params to the reader:
the CLI, and the componentized entrypoints that go through
`DatabaseURLSettings.apply_to_env`.
Only pool and timeout params are inherited, through a single allowlist both
paths share. Anything that decides which tables a query resolves against
stays on the writer, including entries smuggled in through
`database_extra_connection_params`, so a writer `search_path` cannot repoint
reader queries. Params the operator pinned on the replica URL still win.
Co-authored-by: Yassin Kortam <yassin.kortam@gmail.com>
Co-authored-by: yassin <yassin@berri.ai>
Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Cognition serves an OpenAI-compatible /v1/chat/completions endpoint, so it has been onboarded as
custom_llm_provider: openai. That books its traffic as OpenAI, which means OpenAI-specific cost
discounts and provider-level reporting apply to it.
Registers cognition through the JSON provider registry: a providers.json entry with
COGNITION_API_KEY and COGNITION_API_BASE, LlmProviders.COGNITION, the constants.py provider lists,
cost map entries for swe-1.6 and swe-1.7, the provider endpoints matrix, the dashboard provider
fields, and tests. JSON providers can now also be resolved from their base url alone, so an
api_base pointing at a known provider no longer falls through to an unresolved provider.
* test: point the live web search, groq and vertex image suites at models that still exist
Three CircleCI jobs on the staging-to-main promotion are red because the models
their live suites call have been retired by the providers, not because anything
in litellm changed.
openai/gpt-4o-search-preview now answers "has been deprecated" (its dated id
gpt-4o-search-preview-2025-03-11 carries deprecation_date 2026-07-23), so the
two web search conformance tests and the web search cost tracking test move to
gpt-5-search-api, the current search model. It keeps mode chat,
supports_web_search and a search_context_cost_per_query map, so the cost
assertion still resolves.
groq/llama-3.1-8b-instant reached its deprecation_date of 2026-08-16 and Groq
answers "does not exist or you do not have access to it". It follows
groq/llama-3.3-70b-versatile to groq/openai/gpt-oss-120b, the same replacement
PR #37422 already picked. The proxy config that job boots routes on a */*
wildcard, so no config change is needed.
vertex_ai/imagen-3.0-fast-generate-001 404s with "was not found or your project
does not have access to it". Google retired the whole Imagen family across
Vertex and the Gemini API, so there is no Imagen id left to point at. The class
is removed rather than repointed: Vertex image generation is already covered
live by TestVertexAIGeminiImageGeneration on vertex_ai/gemini-2.5-flash-image,
and the Imagen request and response transformations keep their offline coverage
in tests/test_litellm/llms/vertex_ai/image_generation/.
Only live call sites move. Remaining references to the old ids sit in offline
cost-map and transformation tests, where the string is a lookup key and no
request leaves the process.
* chore(lint): ratchet the TQ005 ceiling down to the count this branch reached
Removing the retired TestVertexImageGeneration class cleared one TQ005
violation, so the gate demands the limit come down with it.
make lint-budget-update only lowers a limit by the delta a branch cleared, and
this ceiling already sat 2 above the base count, so the tool landed on 2834
while the gate wants the limit at or below the 2832 this branch reached. The
remaining 2 are that stale headroom, which is exactly what the gate is asking
to reclaim.
* test(ci): serve /moderations from the canned OpenAI mock
The otel proxy E2E job points its `openai/*` wildcard deployment at the
canned mock, and #37492 made `get_model_list` agree with
`get_available_deployment` on bare model names. /moderations now resolves
`omni-moderation-latest` to that wildcard deployment the way
/chat/completions already did, so the request lands on the mock, which
never implemented the route and answers a bare 404.
Add /moderations and /v1/moderations to the mock, returning an
OpenAI-shaped response with one result per input item.
* style(ci): annotate the new moderations locals as Final
The monitors are separate servers with their own password, so the data node's Entra or
IAM token has no standing there. Dropping the provider only when a Sentinel password was
configured left it in place for unauthenticated monitors, where redis-py sends it as an
AUTH the monitor rejects and async Sentinel discovery fails.
The generic container handler returned response.content for endpoints marked
returns_binary before it ran any status or error check, so a non-2xx answer
from the provider was handed back to the caller as raw bytes. Asking for the
content of a container file that does not exist returned the provider's 404
error body as an opaque payload instead of raising.
Move the check ahead of the binary short-circuit and apply it to every
container file endpoint, falling back to the response text when the error body
is not JSON.
A JWKS fetch had no retry, so a single connect timeout to the identity provider
failed authentication outright, and once the cached copy expired there was
nothing to fall back on. How that surfaced depended on the outage shape:
httpx.ConnectTimeout was missing from DB_CONNECTION_ERROR_TYPES so it fell
through to the generic auth handler as a 401 with an empty detail, while a read
timeout took the database path and reported a healthy database as unreachable.
Transport failures are now retried three times with a short backoff, and the
last-known-good JWKS stays usable for a bounded window past public_key_ttl.
That window is public_key_stale_ttl, a new config field defaulting to 3600s and
settable to 0 to fail closed. It is checked on every read against the current
setting rather than baked into the cache entry when it is written, so lowering
it binds immediately instead of waiting for entries written under the old value
to age out, which matters because a shared cache survives the restart an
operator performs to make the change take effect. A copy whose write time
cannot be established is not servable. Only httpx.TransportError unlocks the
stale copy, so an identity provider that answers at all, including with a
narrowed key set, revokes on the next refresh. Every stale serve logs the kid
it authenticated, how long ago that copy was refreshed, and how long until it
stops being trusted.
A sustained outage is remembered for 30s per key url, so it costs one fetch per
window instead of three timeouts per request serialised behind the refresh lock.
Non-200 JWKS responses now raise instead of being cached as the key set, which
previously let an error body overwrite the last-known-good copy. An unreachable
identity provider with no cached copy left returns 503 auth_provider_unavailable.
Resolves LIT-5524
Co-authored-by: Yassin Kortam <yassin@berri.ai>
A request can name more than one model, through a comma-separated model or target_model_names on
the batch and fine-tuning routes, and the gate only looked at the string case, so an unpriced model
riding alongside a priced one went through and billed. Check every candidate and name the unpriced
ones in the 403
Aliases had the same problem on the other side: a group that prices itself through its model_info
block lands in the cost map under its deployment id, and the explicit-cost check walked the raw
model list by group name, so an alias pointing at that group read as unpriced. Resolve the group
through the router the way the pricing check already does
Also correct the 403 copy. Providers that return their own usage cost still bill for these models,
so the accurate claim is that litellm has no pricing of its own for them
The three helpers added for the incomplete-response tests took untyped
parameters, which the repo's typing rule does not allow. Annotate them
through a TYPE_CHECKING block so the runtime imports stay inside the
function bodies like the rest of this file.
Every pod and uvicorn worker schedules its own CheckBatchCost poller against the
shared managed-object table, so two of them can select the same completed batch in
one polling window and both write an aretrieve_batch spend log for it, counting
that batch's cost twice.
Claim the row with a compare-and-swap on batch_processed, and skip the batch when
another pod already holds it. The claim sits immediately before the spend log is
written rather than before the results fetch, because batch_processed is also what
blocks deletion of the files the fetch reads and what keeps an unbilled row
selectable by later poll cycles, so claiming up front would strand the spend of any
worker that died mid-fetch. A failed spend log write hands the row back.
Co-authored-by: Yassin Kortam <yassin@berri.ai>
Some IdPs, ADFS among them, return only `sub` from UserInfo and put the real
identity claims in the ID token or the access token. Those users land in the
Admin UI with no username, email, groups or teams.
Adds an opt-in `GENERIC_INCLUDE_TOKEN_CLAIMS` that merges token claims into the
UserInfo response before the existing `GENERIC_USER_*_ATTRIBUTE` mappings run.
Precedence is UserInfo, then id_token, then access token, and it applies to both
the PKCE and non-PKCE login flows. With the flag unset, behavior is unchanged.
Co-authored-by: Yassin Kortam <yassin@berri.ai>
OpenAI documents computer_use as a supported tool for Daybreak Blue and its
default snapshot gpt-5.6-sol, but neither entry carried supports_computer_use.
Sibling gpt-5.6-cyber and daybreak-red-latest already set it, so /model/info
and the capability gates reported blue as unable to use computer tools.
The gap came in with the source PR rather than the consolidation: #37029 sets
the flag on cyber and red only. Pinned by a new metadata test covering the
daybreak family and the blue alias agreeing with its snapshot.
Adds a chat_completions route module to litellm-core, mirroring the messages
route, plus Anthropic Messages and Bedrock Converse provider configs. The
per-model `rust: true` opt-in now covers /chat/completions for both providers.
The core accepts an allowlisted subset (text conversations, non-streaming) and
returns CoreError::Unsupported for anything else, so tool calls, multimodal
content and streaming fall back to the Python path transparently.
Resolves LIT-5698
redis-py awaits a redis_connect_func that is a coroutine function, so
dropping every connect func the async paths cannot convert took away an
auth path that worked.
A semantic cache lookup embeds the prompt before the request reaches the LLM,
and that embedding call carried no deadline of its own. It inherited the 6000s
request timeout and the Router's num_retries, so an embedding endpoint that is
down or unroutable parked every proxied request for minutes and gave back
nothing but x-litellm-semantic-similarity 0.0 once it finally gave up.
The lookup now runs on its own short deadline, 5s by default, with retries off
so failures cannot stack. Redis, Valkey and qdrant all pick it up, and the
deadline is settable per cache with semantic_cache_embedding_timeout or
globally with SEMANTIC_CACHE_EMBEDDING_TIMEOUT_SECONDS.
tiktoken's BPE merge loop is quadratic in the length of a single regex piece, so a long
run of one repeated character turns a multi-MB payload into minutes of CPU. Encoding in
bounded chunks makes that linear, at a drift of at most ~1 token per chunk boundary.
Chunking alone only makes the stall shorter, so the async paths now count in a worker
thread: tiktoken releases the GIL for its Rust encode, so the loop keeps serving other
requests while a count is in flight. The /utils/token_counter endpoint awaits the new
atoken_counter, and the router's async deployment selection counts off-loop and hands
the result to _pre_call_checks instead of making it count inline.
The chunk size knob is bounded to [1, 4096]: a non-positive value used to raise or
silently report zero tokens, and an arbitrarily large one restored the quadratic cost
this exists to remove. Out-of-range and unparseable values warn and fall back to 1024.
Co-authored-by: Yassin Kortam <yassin@berri.ai>
Budget reservation tokenized every request twice, once for the max-cost
estimate and once for the input-cost estimate, and again per pricing
candidate. Tokenizing is O(prompt) and ran inline, so admitting one large
request stalled every other request the worker was serving.
Count the input tokens once per request and reuse the counts for both
estimates. Prompts above 30K characters of input text are counted in a
worker thread so the event loop stays free. The size heuristic renders the
body rather than walking its values, so tool-schema property names count
toward the threshold, and it sizes every field the counter tokenizes,
tool_choice included.
Co-authored-by: Yassin Kortam <yassin@berri.ai>
The shared logging ThreadPoolExecutor uses an unbounded work queue, so
sync callbacks that fall behind request arrival pin every queued payload
in memory until the task restarts. Cap queued-plus-running work with a
semaphore, shed submissions past the cap, and warn at most once every 30
seconds naming the knob that raises it. No caller of the shared executor
reads the returned future, so shedding is safe.
Co-authored-by: Yassin Kortam <yassin@berri.ai>
Agent registry CRUD (/v1/agents*) sat in agent_routes, which feeds
llm_api_routes, so DISABLE_LLM_API_ENDPOINTS returned "LLM API routes are
disabled for this instance." for every Admin UI Agents tab call. Split the
group the same way MCP is split: agent_inference_routes stays on the data
plane, agent_management_routes joins management_routes, and agent_routes
remains their union for keys configured with allowed_routes=["agent_routes"].
Non-admin callers reached agent CRUD through llm_api_routes before, so the
management paths also join self_managed_routes and the llm_api_routes virtual
key carve-out; the handlers already scope reads by role and 403 non-admin
writes.
Both new groups are tuples, so check_route_access now takes a Sequence and
matches wildcards through a generator instead of materializing an
intermediate list on every call.
A dropped connection anywhere in the budget reset tick used to abort the whole
phase, so every due key, user, team and budget tier stayed unreset until the
next tick ten minutes later. Route the job's DB calls through
call_with_db_reconnect_retry so a transport blip costs one reconnect instead.
Reads replay on any transport error, since re-running a SELECT has nothing to
double-apply. Writes are non-idempotent, a reset assigns spend = 0
unconditionally, so they narrow to DB_RETRY_SAFE_ERROR_TYPES: only a
ConnectError proves the statements never reached the database. A post-send
error like ReadError or ReadTimeout leaves the commit outcome unknown, and
replaying one that already landed would erase whatever was spent since, so
those keep the pre-existing behaviour of failing the tick.
Resolves LIT-5372
Co-authored-by: Yassin Kortam <yassin@berri.ai>
Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
