scriptoriumadmin/litellm
1
0
Fork 0
mirror of https://github.com/BerriAI/litellm.git synced 2026-11-10 08:28:25 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 108
litellm/tests/unit/llms
History
norkelbyte 3b35ae340d
fix(rate-limit): estimate embedding image and audio data URIs as media in TPM pre-call checks (#49831) 
* fix(rate-limit): estimate embedding image and audio data URIs as media in TPM pre-call checks

* fix(rate-limit): count embedding media as media only on models that embed it

A data URI sent to a text-only embedding model is read as text, so it keeps
the text estimate there. On models flagged supports_embedding_image_input,
image and audio use the chat media estimates and PDF and video count at most
the model's max_input_tokens. The same estimate now feeds project and key
limits, the usage-based, latency and cost routers' deployment tpm, and the
deployment itpm check.

* fix(proxy): count embedding media only where the backend reads it as media

Vertex AI, Gemini, Azure AI and Bedrock count a strict base64 data URI as
media per item. Cohere counts it as media only when the whole batch is
media, and Voyage, OCI and unknown models keep the text estimate. File
blocks carrying a data URI count like the bare URI.

The router context window pre-call check and the batch limiter now use the
same estimate, resolving each deployment or JSONL row model to its
backends. The cost map gains supports_embedding_image_input on
voyage/voyage-multimodal-3 and oci/cohere.embed-multilingual-image-v3.0.

* fix(router): hold each embedding media item to the context window on its own

A PDF or video data URI alone already reserves at most the model's
max_input_tokens, but the context window pre-call check then ran the
blanked input through the Responses transform and added a few framing
tokens, so a PDF that Vertex accepts got a 400. When every input item is
media, the check now returns the largest item's media estimate, since an
embedding request embeds each item separately

* fix(proxy): leave embedding media out of the context check and narrow Bedrock media scope

The router context window check now counts only the text of an embedding input, so Titan image lists no longer 400. Bedrock Marengo and Nova with a PDF keep the text estimate since they read those inputs as text or reject them, async_invoke Nova counts images as media, and the unused batch limiter change is dropped.

* fix(bedrock): bill Cohere embeddings by Bedrock's input token count header

* fix(embeddings): check data URI base64 in place and keep Bedrock Cohere text usage

The strict base64 check now scans the payload in place instead of decoding
and re-encoding a full copy. Bedrock's billed token header is used only for
image requests, so text-only Bedrock Cohere usage matches main. The deployment
itpm estimator change is reverted, since the router never passes embedding
input to that check.

---------

Co-authored-by: mateo-berri <277851410+mateo-berri@users.noreply.github.com>
2026-11-03 22:06:59 +00:00
..
a2a fix(a2a): send agent litellm_params.api_key as bearer auth on the A2A hop (#49474) 2026-10-31 12:06:52 -07:00
aiml chore(typing): clear 670 basedpyright Any errors by validating provider payloads at the boundary (#45388) 2026-10-13 15:35:23 -07:00
aiohttp_openai fix(aiohttp_openai): forward service_tier in transform_response (#47263) 2026-10-28 13:14:44 -07:00
anthropic fix(anthropic)!: carry prompt_cache_breakpoint to cache_control on chat completions to Claude (#49783) 2026-11-03 12:56:22 +00:00
apiserpent test(unit): add package markers to migrated unit test directories 2026-09-20 11:53:14 +00:00
aws_polly fix(aws_polly): read the engine from the bare model name (#47963) 2026-10-23 18:10:51 -07:00
azure fix(cache_control)!: keep one deployment's prompt-cache opt-in from reaching siblings on the same model (#49685) 2026-11-03 07:55:51 +00:00
azure_ai fix(model_prices): consolidate verified provider metadata and pricing (#49361) 2026-11-03 11:32:12 -08:00
base_llm feat(decisions): add Parallel AI web-grounded decisions (#49188) 2026-11-03 04:47:02 +00:00
baseten fix(aiand)!: show ai&'s name and logo, and sync ai&'s model list (#49158) 2026-10-30 10:55:26 -07:00
bedrock fix(rate-limit): estimate embedding image and audio data URIs as media in TPM pre-call checks (#49831) 2026-11-03 22:06:59 +00:00
bedrock_mantle fix(bedrock)!: give Invoke and Mantle Claude the model's own default max_tokens (#49692) 2026-11-02 11:42:45 -08:00
bigmodel feat(zai): support response_format and max_completion_tokens, add glm-5v-turbo pricing (#48335) 2026-10-29 20:30:50 -07:00
black_forest_labs refactor(types): replace Any with proven types in 137 files (#44478) 2026-10-05 11:01:45 -07:00
brave chore(typing): clear 670 basedpyright Any errors by validating provider payloads at the boundary (#45388) 2026-10-13 15:35:23 -07:00
bytez test(unit): make every tests/unit directory a package so pytest collection is unique 2026-09-20 11:50:59 +00:00
cerebras test(unit): make every tests/unit directory a package so pytest collection is unique 2026-09-20 11:50:59 +00:00
chat fix(router): read the first stream chunk before returning so a failure before any token runs through the retry and fallback budget (#46493) 2026-10-14 14:19:25 -07:00
chatgpt fix(streaming): keep merged think tags balanced and split SSE lines only on CR/LF (#48755) 2026-10-28 17:00:29 -07:00
clarifai refactor(types): replace Any with proven types in 137 files (#44478) 2026-10-05 11:01:45 -07:00
claude_code refactor(types): replace Any with proven types in 137 files (#44478) 2026-10-05 11:01:45 -07:00
clf_ai_gateway refactor(router): route the router family through RoutingContext and close the outside surface (#49118) 2026-10-30 19:18:54 -07:00
cloudflare fix(streaming): keep merged think tags balanced and split SSE lines only on CR/LF (#48755) 2026-10-28 17:00:29 -07:00
codestral fix(streaming): keep merged think tags balanced and split SSE lines only on CR/LF (#48755) 2026-10-28 17:00:29 -07:00
codex refactor(types): replace Any with proven types in 137 files (#44478) 2026-10-05 11:01:45 -07:00
cohere fix(rate-limit): estimate embedding image and audio data URIs as media in TPM pre-call checks (#49831) 2026-11-03 22:06:59 +00:00
cometapi chore(typing): clear 670 basedpyright Any errors by validating provider payloads at the boundary (#45388) 2026-10-13 15:35:23 -07:00
compactifai refactor(typing): clear 1181 basedpyright Any errors across 431 files (#47789) 2026-10-24 11:31:09 -07:00
crusoe fix(crusoe): update base URL to current endpoint (#33121) 2026-10-15 19:10:58 -07:00
custom_httpx feat(anthropic): pick the beta allowlist row per deployment and surface input_transformations (#49689) 2026-11-03 01:55:26 +00:00
dashscope feat(dashscope): add image edits and text-to-speech (#48998) 2026-10-29 17:48:38 -07:00
databricks fix(mcp)!: answer 400 for a hosted MCP tool a translated route cannot run, drop it under drop_params (#48931) 2026-11-02 13:13:03 -08:00
dataforseo chore(typing): clear 670 basedpyright Any errors by validating provider payloads at the boundary (#45388) 2026-10-13 15:35:23 -07:00
datarobot test(unit): make every tests/unit directory a package so pytest collection is unique 2026-09-20 11:50:59 +00:00
deepagents feat: add litellm.agent() to run claude code, codex, opencode and deep agents through the ai gateway (#43885) 2026-10-01 22:27:49 +00:00
deepgram fix(cost): add deepgram/streaming/nova-3-medical and update stale Nova-3 streaming rates (#49733) 2026-11-02 14:17:10 -08:00
deepinfra fix(openai): stop stripping cache_control from the caller's own messages and tools (#35602) 2026-10-31 00:29:10 -07:00
deepseek fix(mcp)!: answer 400 for a hosted MCP tool a translated route cannot run, drop it under drop_params (#48931) 2026-11-02 13:13:03 -08:00
docker_model_runner test(unit): make every tests/unit directory a package so pytest collection is unique 2026-09-20 11:50:59 +00:00
duckduckgo chore(typing): clear 670 basedpyright Any errors by validating provider payloads at the boundary (#45388) 2026-10-13 15:35:23 -07:00
e2b refactor(types): replace Any with proven types in 137 files (#44478) 2026-10-05 11:01:45 -07:00
edenai test(router_strategy): keep foreign asyncio task noise out of the budget limiter caplog asserts (#43375) 2026-10-29 06:22:06 -07:00
elevenlabs fix(elevenlabs): map wav and pcm TTS formats to 24kHz output formats (#48455) 2026-10-26 13:42:01 -07:00
exa_ai fix(exa): return highlights, highlight scores, score and the response envelope from Exa search (#47423) 2026-10-19 15:45:56 -07:00
fal_ai fix(logging): warn when drop_params drops a requested reasoning level on the remaining paths (#48933) 2026-10-30 08:48:43 -07:00
fastcrw fix(llms): sanitize trailing slashes in provider get_complete_url (#42887) 2026-10-15 02:59:46 -07:00
featherless_ai refactor(llms): expose public names for private provider helpers (#45037) 2026-10-07 08:49:25 -07:00
firecrawl fix(llms): sanitize trailing slashes in provider get_complete_url (#42887) 2026-10-15 02:59:46 -07:00
fireworks_ai fix(fireworks): keep developer items in place for models that accept a later system message (#49399) 2026-11-02 18:46:30 +00:00
friendliai fix(friendliai): drop litellm-internal message fields before sending (#45092) 2026-10-20 01:22:07 +00:00
gdc test: move tests/test_litellm/llms into tests/unit/llms (#43191) 2026-09-25 12:43:23 -07:00
gemini fix(gemini)!: answer 400 for an unsupported chat audio format, drop it under drop_params (#49423) 2026-11-02 04:33:19 -08:00
gigachat fix(mcp)!: answer 400 for a hosted MCP tool a translated route cannot run, drop it under drop_params (#48931) 2026-11-02 13:13:03 -08:00
github_copilot fix(github_copilot): treat a malformed api-key.json like a missing one (#44723) 2026-10-29 03:46:44 -07:00
google_pse chore(typing): clear 670 basedpyright Any errors by validating provider payloads at the boundary (#45388) 2026-10-13 15:35:23 -07:00
gradient_ai fix(llms): sanitize trailing slashes in provider get_complete_url (#42887) 2026-10-15 02:59:46 -07:00
groq fix(openai): stop stripping cache_control from the caller's own messages and tools (#35602) 2026-10-31 00:29:10 -07:00
heroku fix(llms): sanitize trailing slashes in provider get_complete_url (#42887) 2026-10-15 02:59:46 -07:00
hosted_vllm feat(hosted_vllm)!: keep vLLM's stop_reason and token_ids on chat choices and stream chunks (#48393) 2026-11-02 13:26:29 -08:00
huggingface fix(embeddings): honor the caller's timeout on Cohere sync and HuggingFace embedding requests (#48593) 2026-11-02 22:29:37 -08:00
hyperbolic refactor(llms): expose public names for private provider helpers (#45037) 2026-10-07 08:49:25 -07:00
inception refactor(llms): expose public names for private provider helpers (#45037) 2026-10-07 08:49:25 -07:00
infinity chore(typing): clear 670 basedpyright Any errors by validating provider payloads at the boundary (#45388) 2026-10-13 15:35:23 -07:00
jina_ai fix(rerank): preserve token usage in spend logs (#49308) 2026-10-30 13:46:27 -07:00
lambda_ai test: move the unit half of 126 mixed legacy files into tests/unit (#45090) 2026-10-07 14:07:43 -07:00
langflow fix(ci): stop stale CI reds, keep unit tests off the host env, retry CyberArk policy conflicts (#43294) 2026-09-26 09:25:13 -07:00
langgraph fix(streaming)!: fail a stream that ends without a finish reason (#47665) 2026-10-31 11:29:26 -07:00
laya feat: add Bespoke Nimble gateway and OSS classifier support (#44246) 2026-10-02 19:37:08 -07:00
lemonade chore(typing): clear 670 basedpyright Any errors by validating provider payloads at the boundary (#45388) 2026-10-13 15:35:23 -07:00
linkup fix(llms): sanitize trailing slashes in provider get_complete_url (#42887) 2026-10-15 02:59:46 -07:00
litellm_proxy fix(proxy): treat empty agent and skill search embedding model as not configured (#49485) 2026-10-31 15:32:26 -07:00
llamafile refactor(llms): expose public names for private provider helpers (#45037) 2026-10-07 08:49:25 -07:00
lm_studio refactor(llms): expose public names for private provider helpers (#45037) 2026-10-07 08:49:25 -07:00
manus test: migrate SDK and proxy-booting legacy CircleCI tests to their tiers (#45758) 2026-10-11 00:17:21 -07:00
meta fix(realtime)!: keep turn_detection.type when a transcript guardrail disables auto-response (#44939) 2026-10-16 15:24:40 +00:00
meta_llama refactor(llms): expose public names for private provider helpers (#45037) 2026-10-07 08:49:25 -07:00
microsoft_365_copilot refactor(router): route the router family through RoutingContext and close the outside surface (#49118) 2026-10-30 19:18:54 -07:00
milvus test: migrate legacy unit dirs and unrun folders into tests/unit, tests/integration and tests/e2e (#45940) 2026-10-11 06:24:57 -07:00
minimax feat(minimax): add MiniMax-H3 and H3-Max video generation via /v1/videos (#43705) 2026-10-15 05:48:36 -07:00
mistral fix(openai): stop stripping cache_control from the caller's own messages and tools (#35602) 2026-10-31 00:29:10 -07:00
modelscope fix(llms): sanitize trailing slashes in provider get_complete_url (#42887) 2026-10-15 02:59:46 -07:00
mongodb test(unit): make every tests/unit directory a package so pytest collection is unique 2026-09-20 11:50:59 +00:00
moonshot fix(llms): sanitize trailing slashes in provider get_complete_url (#42887) 2026-10-15 02:59:46 -07:00
morph feat(morph): support open models and cached input pricing (#48676) 2026-10-27 18:06:14 -07:00
nadir test: move tests/test_litellm/llms into tests/unit/llms (#43191) 2026-09-25 12:43:23 -07:00
nebius test: move tests/test_litellm/llms into tests/unit/llms (#43191) 2026-09-25 12:43:23 -07:00
neosantara refactor(llms): expose public names for private provider helpers (#45037) 2026-10-07 08:49:25 -07:00
nimble test: move whole-unit legacy test files into tests/unit and delete dead skips (#44807) 2026-10-07 05:02:28 +00:00
nlp_cloud chore(typing): clear 670 basedpyright Any errors by validating provider payloads at the boundary (#45388) 2026-10-13 15:35:23 -07:00
novita test(unit): make every tests/unit directory a package so pytest collection is unique 2026-09-20 11:50:59 +00:00
nscale test(unit): make every tests/unit directory a package so pytest collection is unique 2026-09-20 11:50:59 +00:00
nvidia_nim fix(openai): send chat completions through client.post so the SDK's per-message param transform stops holding the event loop (#46967) 2026-10-16 18:06:01 -07:00
nvidia_riva feat(nvidia_riva)!: add riva_offline mode for offline-only ASR models (#41021) 2026-10-20 23:03:56 +00:00
oci fix(oci): preserve all Gemini system instructions (#48813) (#49458) 2026-10-31 17:49:53 +00:00
ocr test(unit): make every tests/unit directory a package so pytest collection is unique 2026-09-20 11:50:59 +00:00
ollama fix(mcp)!: answer 400 for a hosted MCP tool a translated route cannot run, drop it under drop_params (#48931) 2026-11-02 13:13:03 -08:00
oobabooga fix(streaming): keep merged think tags balanced and split SSE lines only on CR/LF (#48755) 2026-10-28 17:00:29 -07:00
openai fix(bedrock)!: fail a Nova Sonic realtime session whose Bedrock stream drops instead of closing 1000 (#49170) 2026-11-03 07:56:23 +00:00
openai_like fix(cost): split a provider-reported total between input and output by the cost map rate ratio (#49357) 2026-10-31 09:04:50 -07:00
opencode fix(anthropic)!: carry prompt_cache_breakpoint to cache_control on chat completions to Claude (#49783) 2026-11-03 12:56:22 +00:00
openrouter fix(streaming): preserve served models in OpenRouter streams (#49498) 2026-10-31 16:12:51 -07:00
opensandbox chore(typing): clear 670 basedpyright Any errors by validating provider payloads at the boundary (#45388) 2026-10-13 15:35:23 -07:00
opensearch feat(vector_stores): add OpenSearch k-NN search as a vector store provider (#47759) 2026-10-29 08:15:05 -07:00
ovhcloud chore(typing): clear 670 basedpyright Any errors by validating provider payloads at the boundary (#45388) 2026-10-13 15:35:23 -07:00
parallel_ai feat(decisions): add Parallel AI web-grounded decisions (#49188) 2026-11-03 04:47:02 +00:00
parasail refactor(llms): expose public names for private provider helpers (#45037) 2026-10-07 08:49:25 -07:00
pass_through fix(guardrails): exclude litellm_metadata from pass-through guardrail scan text (#48276) 2026-10-26 12:07:58 -07:00
perplexity fix(search): forward Perplexity search filters (#49344) 2026-10-30 17:08:32 -07:00
pg_vector fix(security): remove the publicly known master key from the repo (#44718) 2026-10-06 10:55:24 -07:00
pi feat(harness): add pi harness for litellm.agent() (#45467) 2026-10-20 09:01:40 +00:00
publicai test: migrate tests/llm_translation into unit, integration and e2e tiers (#45759) 2026-10-11 07:58:53 +00:00
ragflow test: migrate legacy unit dirs and unrun folders into tests/unit, tests/integration and tests/e2e (#45940) 2026-10-11 06:24:57 -07:00
recraft refactor(types): replace Any with proven types in 137 files (#44478) 2026-10-05 11:01:45 -07:00
reducto test: move tests/test_litellm/llms into tests/unit/llms (#43191) 2026-09-25 12:43:23 -07:00
replicate chore(typing): clear 670 basedpyright Any errors by validating provider payloads at the boundary (#45388) 2026-10-13 15:35:23 -07:00
runwayml feat(dashscope): add image edits and text-to-speech (#48998) 2026-10-29 17:48:38 -07:00
s3_vectors refactor(router): move router callers onto the entry points and invert the proxy imports (#48847) 2026-10-29 17:26:33 +00:00
sagemaker fix(router)!: keep a deployment's own model_id over the request body's (#49045) 2026-10-30 02:40:01 -07:00
sail test(router_strategy): keep foreign asyncio task noise out of the budget limiter caplog asserts (#43375) 2026-10-29 06:22:06 -07:00
sambanova chore(typing): clear 670 basedpyright Any errors by validating provider payloads at the boundary (#45388) 2026-10-13 15:35:23 -07:00
sap fix(anthropic)!: send /v1/messages PDF documents to OpenAI-shaped providers as file parts (#47166) 2026-10-29 08:06:12 -07:00
scaledown feat(scaledown): add native text provider support (#44168) 2026-10-10 14:01:17 -07:00
scaleway test: migrate tests/llm_translation into unit, integration and e2e tiers (#45759) 2026-10-11 07:58:53 +00:00
searchapi chore(typing): clear 670 basedpyright Any errors by validating provider payloads at the boundary (#45388) 2026-10-13 15:35:23 -07:00
searxng chore(typing): clear 670 basedpyright Any errors by validating provider payloads at the boundary (#45388) 2026-10-13 15:35:23 -07:00
serper chore(typing): clear 670 basedpyright Any errors by validating provider payloads at the boundary (#45388) 2026-10-13 15:35:23 -07:00
snowflake fix(anthropic)!: send /v1/messages PDF documents to OpenAI-shaped providers as file parts (#47166) 2026-10-29 08:06:12 -07:00
soniox feat(soniox): add realtime speech-to-text on /v1/realtime (#44822) 2026-10-29 07:58:38 -07:00
stability refactor(types): replace Any with proven types in 137 files (#44478) 2026-10-05 11:01:45 -07:00
strands_decider feat(strands_decider): call a Strands Decider on an AgentCore runtime ARN (#45678) 2026-10-09 22:51:19 -07:00
tavily fix(tavily): return raw_content, score, favicon and the response envelope from Tavily search (#48684) 2026-10-27 18:52:06 -07:00
tencent fix(logging): warn when drop_params drops a requested reasoning level on the remaining paths (#48933) 2026-10-30 08:48:43 -07:00
tinyfish test: move whole-unit legacy test files into tests/unit and delete dead skips (#44807) 2026-10-07 05:02:28 +00:00
together_ai fix(responses bridge): round-trip tool_search_call and tool_search_output items between turns (#46444) 2026-10-14 21:40:14 -07:00
tool_loop feat(harness): add Harness.TOOL_LOOP, a minimal in-process tool-calling loop (#44391) 2026-10-03 17:33:01 +00:00
topaz fix(llms): sanitize trailing slashes in provider get_complete_url (#42887) 2026-10-15 02:59:46 -07:00
triton fix(streaming): keep merged think tags balanced and split SSE lines only on CR/LF (#48755) 2026-10-28 17:00:29 -07:00
v0 test: move the unit half of 126 mixed legacy files into tests/unit (#45090) 2026-10-07 14:07:43 -07:00
valkey test: migrate legacy provider tests to tests/unit (wave 2, phase 13) 2026-09-20 14:18:49 +00:00
vercel_ai_gateway refactor(llms): expose public names for private provider helpers (#45037) 2026-10-07 08:49:25 -07:00
vertex_ai fix(gemini): resend once with the placeholder signature when the other backend rejects a replayed thought signature (#49697) 2026-11-03 01:24:55 +00:00
vllm fix(vllm): list hosted_vllm wildcard models when the server has no api key (#40066) 2026-11-03 04:13:28 +00:00
volcengine fix(openai): send chat completions through client.post so the SDK's per-message param transform stops holding the event loop (#46967) 2026-10-16 18:06:01 -07:00
voyage fix(llms): sanitize trailing slashes in provider get_complete_url (#42887) 2026-10-15 02:59:46 -07:00
wandb test: move tests/test_litellm/llms into tests/unit/llms (#43191) 2026-09-25 12:43:23 -07:00
watsonx fix(openai): stop stripping cache_control from the caller's own messages and tools (#35602) 2026-10-31 00:29:10 -07:00
xai fix(openai): stop stripping cache_control from the caller's own messages and tools (#35602) 2026-10-31 00:29:10 -07:00
xinference test: move whole-unit legacy test files into tests/unit and delete dead skips (#44807) 2026-10-07 05:02:28 +00:00
you_com test: migrate phase 14 wave 2 provider tests to tests/unit 2026-09-20 11:01:40 +00:00
zai feat(zai): support response_format and max_completion_tokens, add glm-5v-turbo pricing (#48335) 2026-10-29 20:30:50 -07:00
__init__.py test(unit): add package markers to migrated unit test directories 2026-09-20 11:53:14 +00:00
test_api_base_trailing_slash_routes.py fix(llms): join api_base and route with one shared helper (#47535) 2026-10-21 06:59:32 +00:00
test_cache_control_and_reasoning.py test: move tests/test_litellm/llms into tests/unit/llms (#43191) 2026-09-25 12:43:23 -07:00
test_custom_llm.py test: move whole-unit legacy test files into tests/unit and delete dead skips (#44807) 2026-10-07 05:02:28 +00:00
test_file_content_block.py refactor(llms): expose public names for private provider helpers (#45037) 2026-10-07 08:49:25 -07:00
test_file_search_responses.py fix(responses)!: count input tokens the way inference sends the request (#47363) 2026-10-20 15:13:59 +00:00
test_get_complete_url_trailing_slash.py feat(dashscope): add image edits and text-to-speech (#48998) 2026-10-29 17:48:38 -07:00
test_lifecycle_fix.py refactor(llms): expose public names for private provider helpers (#45037) 2026-10-07 08:49:25 -07:00
test_oss_decision.py feat(decisions): add Laya and Bespoke providers to the Add Model form (#45260) 2026-10-14 08:11:11 -07:00
test_polling_url_origin_match.py test: move tests/test_litellm core utils, routing, responses, caching and rust_bridge into tests/unit (#43199) 2026-09-25 17:10:13 -07:00
test_predibase_transformation.py fix(streaming): keep merged think tags balanced and split SSE lines only on CR/LF (#48755) 2026-10-28 17:00:29 -07:00
test_subscription_accounts.py feat(usage): show subscription-covered providers and account fees separately (#44772) 2026-10-30 03:58:36 -07:00