* fix(auth): quiet malformed virtual key rejections to stdout
Reduce noisy invalid-api-key error logs by classifying malformed virtual
keys and routing their rejections to stdout as WARNING instead of stderr
as ERROR. Suppressible via LITELLM_LOG=ERROR or log_client_error_tracebacks=true.
Changes:
- auth_utils: is_invalid_virtual_key_error() classifier and marker functions
- auth_exception_handler: log invalid keys as WARNING to child logger before
identity seeding and callbacks, escalate non-401 transforms to ERROR
- user_api_key_auth: websocket early-raise WebSocketException(1008) to avoid
double-logging at HTTP layer
- _logging: child logger verbose_proxy_stdout_logger with no handler/level;
LevelRoutingStreamHandler routes its WARNING records to stdout; handler
setLevel in _turn_on_json() closes JSON config handler level leak
- test_auth_exception_handler: new test case verifying malformed-key logs
at WARNING with marker retention through transformations
Fixes LIT-5362
* fix(auth): classify malformed-key 401 by raise-site marker, not message text
Review round 1 (Greptile P2, veria Low):
- Move the marker attribute name to litellm/constants.py per the shared
sentinel convention
- Stamp the marker on the malformed-key 401 where it is raised and classify
only by it. Message text is caller-influenceable on other 401s (vector
store ids, organization ids are interpolated into their messages), so a
phrase match would let a request body demote an authorization failure to
the quiet log path
- Regression test: a 401 carrying the phrase but not the marker stays at
ERROR on stderr
At end of drain the pump enqueued the sentinel first and picked the
billing mode from client_detached afterward, so a client that consumed
the sentinel and tore the relay down before the pump resumed (possible
whenever the sentinel enqueue hit a full queue) had its fully delivered
response billed through the teardown path, skipping the proxy's
post-response hook. Bill or park before the sentinel goes out, and let
an unconsumed sentinel fall back to dispatching the parked billing.
Shadow eval jobs previously targeted only virtual keys, so deployments on
pure JWT auth (which present no key at all) could never sample their
traffic. Jobs now carry a typed (target_type, target_id) pair covering
keys, teams, and users; sampling matches the identity every request
resolves to at auth time, so team and user jobs cover JWT traffic with
no client changes.
Resolves LIT-6578
* feat(complexity_router): escalate oversized prompts to a tier that fits before dispatch
The classifier scores complexity and never prompt size, so a long agentic
session whose newest ask is trivial classifies SIMPLE onto a small-window
tier and the provider rejects it with a context-window 400 that nothing
retries. The gate runs after classification on every decision path
(classify tail and session-affinity pin), estimates prompt tokens
including the out-of-band carriers (top-level system, tools,
instructions), and when the decided tier provably cannot hold the prompt
moves the request to the lowest configured tier with a model whose
declared window fits, restricting the pick to fitting models when the
decided tier can keep it. Models with no resolvable window are never
escalated away from or onto, escalated decisions are never written as
session pins, and the decision records context_escalated plus the
original tier in spend logs.
Resolves LIT-6503
* fix(complexity_router): judge groups by smallest window, bound skips by bytes, filter adaptive picks
Review-round rework, one mechanism per finding. A group is judged by its
smallest resolvable deployment window, since the core router picks within
a group with no fit check. The counting skip is gated on UTF-8 byte
length, which BPE token counts can never exceed, so token-dense scripts
cannot slip past it; only a real tokenizer count ever moves a request and
a failed count leaves the placement alone. The fit facts now filter every
adaptive phase including cold start and the tier fallbacks. Window
questions adopt the declared provider and never resolve authenticating
providers, and a router instance without get_model_list degrades the gate
to a no-op. Tests rebuilt on real Router instances resolving deployment
model_info end to end, plus a full-path test through
async_get_available_deployment
The batch start told the seed which LiteLLM_SpendLogs rows were its own, but
using it as a hard cutoff also dropped rows another pod had already persisted.
Those rows are only repaid by that pod's own increment, so if it died first the
window row stayed permanently under the recorded spend.
The seed now reads both sums in one scan and takes off this batch's own spend,
flooring at the pre-batch total for the case where its log rows have not landed
yet. Redis payloads keep an empty request_ids so a leader from before the field
was dropped can still merge what it pops during a rolling deploy.
Claude-Session: https://claude.ai/code/session_01QvQzYztinxj8ZuD5YxbVdL
GigaChat reports prompt_tokens and total_tokens after subtracting cached
tokens (the docs example is prompt_tokens=1, precached_prompt_tokens=37,
total_tokens=5, so the fields are disjoint, not a subset). Map to the
OpenAI convention by adding precached_prompt_tokens back onto prompt and
total while still surfacing it as prompt_tokens_details.cached_tokens.
precached_prompt_tokens is a subset of prompt_tokens (OpenAI cached_tokens
semantics), so map it to prompt_tokens_details.cached_tokens instead of
adding it on top of prompt/total. Emit stream usage from any final chunk
carrying it rather than only finish_reason stop, which dropped tokens for
function_call and length streams. Merge auth metadata into a new dict in
the gigachat router handler instead of mutating the shared parsed-body
cache in place.
* fix(otel): emit cache token counts on OTel v2 LLM spans
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(otel): trim comment in LLMUsage adapter
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(otel): drop casts in LLMUsage cache token adapter
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(deps): bump restrictedpython to 8.3 for GHSA-ffg3-p8fm-mjx2
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>
`ChatCompletionFileObject` is in the union `_count_content_list` accepts, but
`file` was missing from its match, so every local count of a Responses
`input_file` raised `Invalid content item type: file`. On
/v1/responses/input_tokens that surfaced as an opaque 500 whenever the model's
provider counting API refused the block and the local tokenizer took over.
Count it the way the module already counts the same thing in Anthropic's
dialect: the filename like a document title, the inline bytes through the
image pricer.
The Responses-to-chat transform dropped the filename OpenAI requires next to
file_data, so a request carrying an inline PDF counted 13 tokens instead of 36
and a real completion through the chat bridge got a 400.
- sync llm_passthrough_route: read and close an error-status streaming
response before mapping it, so upstream 4xx/5xx surface as the provider
error instead of httpx.ResponseNotRead
- AsyncPassthroughStreamingResponse: expose aiter_bytes() and carry
_hidden_params so the router attaches headers in place instead of
wrapping the stream in HiddenParamsAsyncIteratorWrapper, which 500'd
every streaming azure router-model passthrough request
- logging: swap the passthrough httpx result for the transformed
ModelResponse/EmbeddingResponse when firing success callbacks
- get_llm_provider: resolve gigachat from its api base and drop the dead
gigachat_models elif branch
- constants: register the gigachat api base in openai_compatible_endpoints
Assistant list content was forwarded to /v1/responses/input_tokens as chat
`text` blocks, which the Responses API rejects (it accepts only output_text
and refusal inside an assistant turn). The 400 sent the whole request to the
local tokenizer, so any conversation with an assistant turn silently lost
provider-exact counting, including the image counting added in 73ab647b1c.
Assistant content now collapses to the plain string the Responses API counts
identically, and image parts are kept to user turns where they are legal.
* fix(vertex_ai): graft default vertex path when api_base has a version-only path
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(vertex_ai): keep query and fragment placement when grafting vertex path
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(vertex_ai): merge alt=sse into existing query when streaming
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
---------
Co-authored-by: yassin <yassin@berri.ai>
Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* feat(spend_tracking): persist router metadata in spend logs for internal router models
* test(spend_tracking): expect router_metadata key in exact-payload tests, type the routed-kwargs helper
When the upstream errors while the client is still connected, the pump
forwards the exception through the relay queue so the proxy's failure
handling re-raises it. If the client disconnects before consuming that
queued exception, neither the failure hook nor billing ran and the spend
row was lost. The pump now waits for client detach and, if the exception
was never consumed, salvages partial spend like the post-disconnect
error path.
Also rewrites the bedrock disconnect logging test to the detached-pump
contract: billing fires after the upstream drain completes, not
synchronously at aclose().
The chat-to-Responses reverse transform kept only text blocks, so an image
input was dropped before the count went to OpenAI. A 256x256 image request
counted 13 tokens instead of 268.
/v1/responses/input_tokens returned 200 with a count for an empty
"input" ("" or []), while OpenAI returns a 400 missing_required_parameter.
The route also went through optimistic budget reservation, which is only
released by LLM success/failure callbacks that a token count never
reaches, so every call leaked a reservation until TTL expiry and could
429 real traffic. Both routes plus the /openai alias now join
/utils/token_counter in the reservation exemption set.
