Images nested inside an Anthropic `tool_result` block were dropped when the
request was adapted for an OpenAI-compatible provider, because the OpenAI tool
message shape only carried text. Hoist those images out of the tool result and
into a following user message so the model can still see them, and widen the
tool message content type to accept image parts.
Adds the OCR 3 model (mistral-ocr-2512) released 2025-12-18 to both the
root and bundled backup cost maps at $2 / 1000 pages and $3 / 1000
annotated pages, mirroring the existing Mistral OCR entries. Regresses
the pricing in both maps and verifies completion_cost scales per page.
* feat(mistral): support Mistral OCR 4 (mistral-ocr-4-0)
Add the mistral/mistral-ocr-4-0 model to the cost map and reprice
mistral/mistral-ocr-latest, which now resolves to OCR 4 server-side,
at $4 / 1000 pages. Add the include_blocks param so callers can request
OCR 4's paragraph-level bounding boxes and typed content blocks.
OCR 4's new per-page response fields (blocks, confidence_scores, tables,
hyperlinks, header, footer) already pass through transform_ocr_response
via the extra="allow" config on OCRPage; add a regression test pinning
that behavior alongside cost and param coverage.
* fix(mistral): revert unverified OCR 4 annotation_cost_per_page bump
Mistral's published OCR 4 pricing lists $4/1000 pages for the API and no
separate annotation rate; the $5/1000 figure is the distinct Document AI
(Studio) tier. The earlier 0.003 -> 0.005 bump on annotation_cost_per_page
had no cited source, and ocr_cost() never reads that field (it bills off
ocr_cost_per_page), so the value is documentation-only.
Revert annotation_cost_per_page to the existing 0.003 convention for both
mistral-ocr-latest and mistral-ocr-4-0, keeping only the verified, tested
ocr_cost_per_page: 0.004 change.
* fix(mistral): set OCR 4 annotation_cost_per_page to verified $5/1000 rate
Verified against Mistral's authoritative sources: the pricing page, the
OCR 4 announcement, and the ocr-4-0 model card all list OCR 4 at $4/1000
pages for basic OCR and $5/1000 for annotated pages (Document AI). The
$5/1000 figure is the annotated-pages rate, which is exactly what
annotation_cost_per_page encodes, mirroring the original OCR entry's
0.001 basic / 0.003 annotated split.
Restore annotation_cost_per_page to 0.005 for mistral-ocr-latest and
mistral-ocr-4-0; the earlier revert to 0.003 was based on an incomplete
reading that treated Document AI as a separate product. ocr_cost_per_page
stays 0.004, which is the value billed by ocr_cost().
* fix(mistral-rust): include_blocks in Rust OCR supported params
---------
Co-authored-by: Cursor Agent <cursoragent@cursor.com>
* fix: correct amazon.titan-embed-text-v2 input price to $0.02/1M tokens (#29693)
* fix: correct amazon.titan-embed-text-v2 input price to $0.02/1M tokens
* test: scope local cost map env var with monkeypatch to avoid test pollution
* fix(sensitive_data_masker): fully mask secrets at or below the reveal threshold (#30764)
* fix(sensitive_data_masker): fully mask secrets at or below the reveal threshold
_mask_value did partial reveal by showing the first visible_prefix and last
visible_suffix characters, but for a value whose length was at or below
visible_prefix + visible_suffix (8 by default) it returned the value verbatim.
A value of exactly 8 chars fell through the length guard and computed
masked_length == 0, reconstructing the original string with no mask characters;
anything shorter hit the early return. Either way short credentials were emitted
in plaintext.
mask_dict routes real secrets through this path, so an 8-char-or-shorter redis
password, api key, or token could be written to logs and the UI unmasked. The
sibling helper mask_sensitive_keys already guards this case; _mask_value now does
the same by fully masking any value at or below the threshold.
* fix(sensitive_data_masker): add mask_short_values opt-out for truncation callers
Fully masking short values is the right default for secret masking, but
CooldownCache reuses the masker purely to truncate exception messages to the
first 50 characters, and it relies on short messages being returned readable.
Masking those blanked out short exception text and broke its tests.
Add a mask_short_values flag (default True, secure) and have CooldownCache pass
False so it keeps the truncation behavior, while every secret-masking caller
still gets short values fully masked.
* fix(mcp_debug): opt out of short-value masking to keep diagnostic token preview
MCPDebug uses the masker to preview auth tokens in debug headers and documents
that values of 10 chars or fewer are shown unchanged so token types stay
distinguishable. Pass mask_short_values=False so that diagnostic behavior is
preserved while secret maskers keep masking short values.
* fix(mcp_debug): mask short auth values in debug headers instead of echoing them
Earlier this masker opted out of short-value masking to keep a token preview, but
that echoes short authorization and token values verbatim in debug response
headers, which is the same leak this change is meant to close. Auth material
should never be emitted in full, so mask short values here too; the first/last
character preview still applies to longer tokens. Only CooldownCache keeps the
opt-out, since it truncates exception text rather than masking secrets.
* test(mcp_debug): assert masked short value preserves length
* refactor(fireworks_ai): remove deprecated audio transcriptions endpoint (#30917)
Fireworks AI deprecated audio inference on 2026-06-10
(https://docs.fireworks.ai/updates/changelog#audio-inference-and-image-generation-deprecation).
Live API testing confirms the endpoint is already non-functional: a valid
Fireworks API key receives HTTP 401 "Unauthorized" from
api.fireworks.ai/inference/v1/audio/transcriptions for every request,
regardless of payload. The audio-prod.api.fireworks.ai host referenced in
the test suite returns 401 for every path; the entire host is decommissioned.
Remove the dead FireworksAIAudioTranscriptionConfig class and every
reference to it across the codebase:
- Delete litellm/llms/fireworks_ai/audio_transcription/ directory (17-line
config class that inherited from OpenAIWhisperAudioTranscriptionConfig)
- Remove the Fireworks branch from
ProviderConfigManager.get_provider_audio_transcription_config() in
litellm/utils.py; update the stale comment in
get_optional_params_transcription that referenced fireworks ai
- Remove the FireworksAIAudioTranscriptionConfig entries from
LLM_CONFIG_NAMES and _LLM_CONFIGS_IMPORT_MAP in
litellm/_lazy_imports_registry.py
- Remove the TYPE_CHECKING re-export in litellm/__init__.py
- Remove the transcription branch in the fireworks_ai case of
get_supported_openai_params() in
litellm/litellm_core_utils/get_supported_openai_params.py
- Remove the whisper-v3 and whisper-v3-turbo entries from
model_prices_and_context_window.json and
litellm/model_prices_and_context_window_backup.json (both had
mode: audio_transcription and zero-cost pricing)
- Remove the TestFireworksAIAudioTranscription test class and its
imports from tests/llm_translation/test_fireworks_ai_translation.py
No other provider is affected. The openai_compatible_providers list,
FireworksAIMixin, and the OpenAI Whisper transcription handler all stay
because they are shared with other Fireworks endpoints and other
providers. The provider_endpoints_support.json registry already had
audio_transcriptions set to false for fireworks_ai.
* feat: add darkbloom provider (#30876)
* feat: add darkbloom provider
* fix: document darkbloom provider endpoints
* fix: address darkbloom review feedback
* fix: update darkbloom tool metadata
* fix: fail fast for non-Postgres database URLs (#30883)
* fix(proxy): fail fast on non-PostgreSQL DATABASE_URL instead of hanging on startup
LiteLLM's Prisma datasource is pinned to provider = 'postgresql', so a sqlite:// or mysql:// DATABASE_URL can never connect.
Today that surfaces as an opaque startup stall where the port never binds, and a separate 'DB not connected' 500 on /key/generate when no DATABASE_URL is set at all leaves operators guessing what to configure.
Validate the DATABASE_URL / DIRECT_URL scheme in run_server before any Prisma call and exit with an actionable message naming the unsupported scheme.
Also reword CommonProxyErrors.db_not_connected_error to tell the operator to set DATABASE_URL to a postgresql:// connection string.
Add regression tests covering postgres acceptance and sqlite/mysql/mssql rejection.
* fix: resolve CI failures and proxy DB URL typing issue
* fix(proxy): fail fast on non-PostgreSQL DATABASE_URLs with clear startup errors instead of hanging
* Validate DIRECT_URL alongside DATABASE_URL startup guards
* fix(bedrock): surface modeled HTTP status for mid-stream error events so 5xx is retryable (#24608) (#30946)
* fix(bedrock): surface modeled HTTP status for mid-stream error events (#24608)
* test(bedrock): mid-stream server errors trigger streaming fallback (#24608)
* style(bedrock): black-format stream-error helper (#24608)
* fix(mcp): re-land native tool preservation with typed annotations (#30645)
* fix(mcp): preserve native tools in semantic filter hook with typed annotations
* fix(mcp): tighten _is_mcp_tool Chat Completions shape check
* fix(sambanova): return embeddings supported params instead of dropping them (#30937)
* fix(router): send fallback metadata when streaming (#30914)
When a streaming request triggers a fallback, there was previously no way to
know it happened. This commit addresses this in a few ways:
1. The response now correctly populates the fallback headers
(`x-litellm-attempted-fallbacks`) so callers know a fallback happened.
2. The correct model ID is passed in the streaming chunks.
3. A streaming chunk with the fallback error can be optionally sent back
to the client (opt-in) by passing `include_fallback_errors: true` in
the request.
The format of the fallback errors while streaming is intentionally OpenAI
compatible to not break existing libraries that parse these events. It was
tested with Vercel's AI SDK (ai-sdk.dev). It is also opt-in, so it is not
delieved unexpectedly to callers by default.
* fix(mistral): drop output-only reasoning fields from input messages (#30884)
LiteLLM attaches reasoning_content and thinking_blocks to assistant
responses. Replaying those assistant turns verbatim forwarded the fields
back to Mistral, whose input schema forbids unknown keys, so the whole
request failed with a 422 extra_forbidden and reasoning models became
unusable across multiple turns.
Strip both fields from assistant messages before the request is built, in
a spot that runs ahead of the image/file branch so it applies on every
path. Fixes#30835
Co-authored-by: Cursor <cursoragent@cursor.com>
* fix(perplexity): bill search queries at the per-request price, not 1/1000 of it (#30652)
* fix(perplexity): bill search queries at the per-request price, not 1/1000
The fallback cost calculator divided search_context_cost_per_query by
1000, but that field stores the per-request price in USD: sonar is
{low: 0.005, medium: 0.008, high: 0.012}, matching Perplexity's published
$5/$8/$12 per 1,000 requests expressed per request. The gemini cost
calculator reads the same field per request with no division (its
docstring calls it "the per-request cost").
The division understated search cost by 1000x on every Perplexity call
that falls back to manual calculation (i.e. when the API does not return
a pre-computed usage.cost). Use the value directly.
Update the tests that had encoded the /1000 factor in their expectations,
and drop an unused import flagged by ruff in the touched test file.
* test(perplexity): update integration test search-cost expectations to per-request
The integration tests still encoded the old /1000 search-cost factor, so
they failed once the fallback calculator was corrected to bill
search_context_cost_per_query per request. Update the four expected-cost
computations (and the high-volume dollar-value comments) to match.
* test(perplexity): drop unused mock imports flagged by ruff
* fix: include model_access_groups when expanding all-team-models in get_team_models (#30622)
* fix(fireworks_ai): return None for transcription in get_supported_openai_params
Fireworks AI deprecated audio inference on 2026-06-10; the endpoint is
decommissioned. Without an explicit transcription branch, requests with
request_type='transcription' fell through to the else and returned
FireworksAIConfig chat-completion params. Return None instead to signal
the provider does not support transcription.
* fix(proxy): gate include_fallback_errors behind expose_fallback_errors_to_caller setting
Without an operator gate, any authenticated caller could set include_fallback_errors=True,
trigger a fallback, and read raw upstream exception messages from the
x-litellm-fallback-errors header and the litellm-fallback-metadata SSE event.
Strip include_fallback_errors from request data in common_processing_pre_call_logic
when expose_fallback_errors_to_caller is not set, so the router never builds the
error list. Also gate _should_include_fallback_errors on the same setting as a
secondary check for the streaming SSE injection path.
* test(proxy): opt in to expose_fallback_errors_to_caller in streaming SSE test
The operator gate added in e7ff3e1 means include_fallback_errors is only
honoured when general_settings.expose_fallback_errors_to_caller is True.
Set that flag via monkeypatch in the test that exercises the emit path.
* test(prompt_templates): make test_convert_url hermetic instead of hitting picsum.photos
test_convert_url called convert_url_to_base64 against a live picsum.photos
URL and asserted nothing, so it added no real signal and broke CI whenever
the host was unreachable (it was returning 522 and blocking this branch).
Replace the live call with a mocked HTTP client and assert the produced
base64 data URL, so the conversion path is exercised deterministically with
no network dependency. This suite runs under VCR, which is why a transport
level mock (respx) does not reliably intercept; mocking the client object
itself is robust regardless.
* fix(interactions): drop role from Interaction response to match Google spec
Google removed the output-only role field from the Interaction schema (it
now lives only on Turn), so the live OpenAPI compliance canary started
failing with 'role' not in spec. Reconcile our generated types by removing
role from Interaction, CreateModelInteractionParams, CreateAgentInteractionParams
and from the LiteLLM InteractionsAPIResponse/InteractionsAPIStreamingResponse,
stop stamping role=model in the responses-to-interactions transformation, and
update the compliance and integration tests accordingly. Turn.role is kept
since the spec still defines it.
* fix: align all-team-models sentinel access
* fix(router): forward include_fallback_errors through multi-hop fallbacks
run_async_fallback received include_fallback_errors as an explicit named
parameter, so it was bound out of **kwargs and never reached the nested
async_function_with_fallbacks call. Multi-hop fallback chains (a fallback
group that itself fails over) therefore stopped collecting fallback errors
beyond the first hop when a caller opted in. Re-inject the flag into kwargs
before the nested call so inner hops keep accumulating errors, which
add_fallback_headers_to_response already merges across levels.
---------
Co-authored-by: Srivatsa Kamballa <skamb10@uic.edu>
Co-authored-by: Ahmad Shahzad <107808273+shzdehmd@users.noreply.github.com>
Co-authored-by: Jeremy Chapeau <113923302+jychp@users.noreply.github.com>
Co-authored-by: KRISH SONI <67964054+krishvsoni@users.noreply.github.com>
Co-authored-by: Kent <72616338+kingdoooo@users.noreply.github.com>
Co-authored-by: Ayush Shekhar <106994833+ayushh0110@users.noreply.github.com>
Co-authored-by: dav nguyxn <hoangson091104@gmail.com>
Co-authored-by: Tal Marian <tal.marian@island.io>
Co-authored-by: Hemant K <51333870+hemant1026@users.noreply.github.com>
Co-authored-by: Cursor <cursoragent@cursor.com>
Co-authored-by: Yash Raj Pandey <55940078+devYRPauli@users.noreply.github.com>
Co-authored-by: Zang Peiyu <166481866+factnn@users.noreply.github.com>
Co-authored-by: mateo-berri <277851410+mateo-berri@users.noreply.github.com>
Unit Tests: Proxy DB Operations / proxy-db (auth-checks, tests/proxy_unit_tests/test_auth_checks.py tests/proxy_unit_tests/test_user_api_key_auth.py, 20, 8) (push) Waiting to run
Unit Tests: Proxy DB Operations / proxy-db (remaining, tests/proxy_unit_tests --ignore=tests/proxy_unit_tests/test_key_generate_prisma.py --ignore=tests/proxy_unit_tests/test_auth_checks.py --ignore=tests/proxy_unit_tests/test_user_api_key_auth.py, 30, 8) (push) Waiting to run
* feat(redis): add circuit breaker to RedisCache to fast-fail when Redis is down (#24181)
* feat(redis): add circuit breaker env var constants
* feat(redis): add RedisCircuitBreaker and apply guard decorator to all async ops
* fix(dual_cache): fall back to L1 instead of re-raising on Redis increment failures
* test(caching): add circuit breaker unit tests
* fix(redis): fast-fail concurrent HALF_OPEN probes — only one probe at a time
* fix(dual_cache): return None fallback when in_memory_cache is absent and Redis fails
* test(caching): add regression tests for HALF_OPEN concurrency and None fallback
* Fix blocking sync next in __anext__ (#24177)
* Fix blocking sync next
* Update tests/test_litellm/litellm_core_utils/test_streaming_handler.py
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
* fix PEP 479 regression in __anext__ sync iterator exhaustion
asyncio.to_thread re-raises thread exceptions inside a coroutine, where
PEP 479 converts StopIteration to RuntimeError before any except clause
can catch it. Add _next_sync_or_exhausted() module-level helper that
catches StopIteration in the thread and returns a sentinel instead, then
raise StopAsyncIteration in the coroutine.
Also rewrites the non-blocking test to use asyncio.gather() instead of
asyncio.create_task() (which returned None on Python 3.9 / pytest-asyncio
in CI), and adds an exhaustion regression test that drains the wrapper
fully and asserts no RuntimeError leaks out.
---------
Co-authored-by: Emerson Gomes <emerson.gomes@thalesgroup.com>
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
* feat: add git-subdir source type to claude-code/plugins API (#24223)
Support a third plugin source type `git-subdir` alongside the existing
`github` and `url` types, as documented in the official Claude Code
plugin marketplaces spec.
New format: {"source": "git-subdir", "url": "...", "path": "subdir/path"}
- Validates url and path fields are present and non-empty
- Rejects absolute paths, '..' segments, backslashes, and percent-encoded
traversal sequences (including double-encoded variants via regex check)
- Extracts path validation into _validate_git_subdir_path() helper
- Updates Pydantic field description to document all three source types
- Adds isValidUrl() check for url/git-subdir source types in the UI form
- Adds "Git Subdir" option to the UI form with a required Path field
- Adds unit tests covering success, update, missing/empty fields,
path traversal variants, and unknown source type
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
* [FEAT] add extract_header and extract_footer to Mistral OCR supported params (#24213)
* docs: add git-subdir source type to claude-code plugin marketplace docs (#24289)
* fix(ui): swap J/K keyboard navigation in log details drawer (#24279) (#24286)
J should navigate down (next) and K should navigate up (previous),
matching vim/standard conventions.
* fix: use async_set_cache in user_api_key_auth hot path (#24302)
* fix: use async_set_cache in auth hot path to avoid blocking event loop
* test: assert no blocking set_cache call in _user_api_key_auth_builder
* test: broaden blocking call check to all sync DualCache methods
* test: fix regression test to actually catch blocking cache calls
* fix: ruff lint unused variable + UI build MessageManager error
- litellm/caching/redis_cache.py: remove unused variable 'e' in circuit
breaker exception handler (F841)
- add_plugin_form.tsx: use MessageManager.error() instead of undefined
message.error() for git URL validation
Co-authored-by: Ishaan Jaff <ishaan-jaff@users.noreply.github.com>
* docs: add REDIS_CIRCUIT_BREAKER env vars to config_settings reference
Add REDIS_CIRCUIT_BREAKER_FAILURE_THRESHOLD and
REDIS_CIRCUIT_BREAKER_RECOVERY_TIMEOUT to the environment variables
reference table so test_env_keys.py passes.
Co-authored-by: Ishaan Jaff <ishaan-jaff@users.noreply.github.com>
---------
Co-authored-by: Emerson Gomes <emerson.gomes@thalesgroup.com>
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
Co-authored-by: Vincenzo Barrea <manamana88@users.noreply.github.com>
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
Co-authored-by: Robert Kirscht <rkirscht242@gmail.com>
Co-authored-by: Imgyu Kim <kimimgo@gmail.com>
Co-authored-by: Cursor Agent <cursoragent@cursor.com>
Co-authored-by: Ishaan Jaff <ishaan-jaff@users.noreply.github.com>
Fixes#23890 — Mistral's Voxtral transcription with `diarize=true` returns
`segments` (with speaker_id, timestamps) and `language`, but these fields
were dropped when mapping the response to TranscriptionResponse.
Add MistralAudioTranscriptionConfig for Mistral's /v1/audio/transcriptions
endpoint, enabling litellm.transcription() with mistral/voxtral-mini-latest
and other Voxtral models. Supports multipart form-data with OpenAI-compatible
params (language, temperature, response_format, timestamp_granularities)
plus Mistral-specific params like diarize.
* Enhance Mistral API: Add support for parallel tool calls and refine name handling in tool messages. Plus, introduce a new test for parallel tool calls in the Mistral model.
* tests
* make mypy happy
* Refine name handling in Mistral chat transformation: clarify conditions for removing the 'name' field based on message role and content.
* refactor: streamline Mistral integration by removing deprecated references and adding a new handler
- Removed "mistral" from the list of compatible providers in constants.
- Updated the completion function in main.py to utilize the new Mistral handler.
- Deleted outdated Mistral chat and embedding files.
- Introduced a new handler for Mistral chat completions, implementing the llm_http_handler pattern.
- Added integration tests for the Mistral handler to ensure proper API base and key handling.
* lint
* fix: remove unneeded handler object
* add tests
* Addres PR comments
* Enhance Mistral API: Add support for parallel tool calls and refine name handling in tool messages. Plus, introduce a new test for parallel tool calls in the Mistral model.
* tests
* make mypy happy
* Refine name handling in Mistral chat transformation: clarify conditions for removing the 'name' field based on message role and content.
* handle mistral returning '' instead of None
* Enhance Mistral API: Add support for parallel tool calls and refine name handling in tool messages. Plus, introduce a new test for parallel tool calls in the Mistral model.
* tests
* make mypy happy
* Refine name handling in Mistral chat transformation: clarify conditions for removing the 'name' field based on message role and content.
- Updated the `_add_reasoning_system_prompt_if_needed` method to maintain the original format of list content when prepending the reasoning prompt.
- Adjusted tests to verify that both string and list content types are correctly handled, ensuring the reasoning prompt is added without altering the content structure.
- Updated the `_add_reasoning_system_prompt_if_needed` method to convert list content to strings before prepending the reasoning prompt.
- Adjusted tests to verify that system messages with list content are correctly transformed into strings, ensuring original content is preserved.
- Revised the reasoning support indicators in the Mistral model documentation for clarity.
- Improved the `_add_reasoning_system_prompt_if_needed` method to handle both string and list content types for system messages, ensuring the reasoning prompt is correctly prepended.
- Added a new test case to verify the functionality of adding the reasoning system prompt when the existing content is a list.
* Added support for reasoning parameters in magistral models, including "reasoning_effort" and "thinking".
* Updated the MistralConfig class to handle reasoning system prompts.
* Implemented tests to verify reasoning functionality and ensure correct parameter mapping for magistral models.
* Enhanced the model prices JSON to reflect new reasoning capabilities.
