litellm/litellm/types
Mateo Wang 3fec5075fa
feat(typesafe): support Jev 1.14 images, box questions, and long-context headers (#46596)
* feat(typesafe): support Jev 1.14 images, box questions, and long-context headers

* fix(cost): drop the typesafe/jev-1.14.0 entry main already prices

* fix(typesafe): send images only with the images header and box questions only to models that answer them

TypeSafe takes assets and input_image parts only when the request carries
TypeSafe-Danger-Zone-Images: true, so without it they follow drop_params like
any other provider that cannot take images. Box support now comes from a
supports_box_questions cost-map flag set on jev-1.14.0 and jev-preview, so
jev-1.13.0 and jev-latest refuse or drop a box question before the call. A
malformed provider_specific_header answers 400 instead of 500.

* fix(typesafe): allow null box confidence and validate asset references and base64

* fix(typesafe): gate box questions on the images header, drop unreadable assets under drop_params, forward opt-in headers on the pass-through

---------

Co-authored-by: devin-ai-integration[bot] <158243242+devin-ai-integration[bot]@users.noreply.github.com>
2026-10-14 22:29:48 -07:00
..
containers refactor: expose public hidden_params accessors (#44668) 2026-10-07 00:01:33 -07:00
google_genai refactor: expose public hidden_params accessors (#44668) 2026-10-07 00:01:33 -07:00
images
integrations fix(prometheus): skip custom metadata labels a metric already carries (#46267) 2026-10-13 18:20:52 -07:00
interactions refactor: expose public hidden_params accessors (#44668) 2026-10-07 00:01:33 -07:00
litellm_core_utils fix(streaming): keep an explicit provider prompt_tokens=0 or completion_tokens=0 in streamed usage 2026-09-21 14:01:08 -07:00
llms feat(xai): bill grok voice realtime sessions (#46600) 2026-10-15 05:27:02 +00:00
management_endpoints refactor(types): move complexity router request models next to their base config (#45780) 2026-10-10 19:07:41 +00:00
mcp_server feat(mcp): separate confidential client for ID-JAG leg 2 (#46211) 2026-10-14 10:04:31 -04:00
passthrough_endpoints refactor(sdk): import UserAPIKeyAuth from litellm.types.proxy.auth (#45806) 2026-10-10 21:27:48 +00:00
prompts perf(types): defer pydantic schema builds via shared LiteLLMBaseModel (#44720) 2026-10-05 22:10:08 -07:00
proxy feat(access-groups): grant Claude Code skills through access groups (#46294) 2026-10-14 23:18:50 +00:00
repositories feat(usage): add top users by spend and user activity views (#45310) 2026-10-10 12:21:43 -07:00
responses fix(responses): translate OpenAI tool_search to Anthropic tool search for Claude models (#45882) 2026-10-14 20:42:22 -07:00
secret_managers feat(secret_managers): support customer-managed KMS key for virtual keys stored in AWS Secrets Manager (#40475) 2026-09-11 09:54:08 -07:00
videos feat(dashscope): add Wan 3.0, Wan 2.7 and HappyHorse video generation via /v1/videos (#43706) 2026-10-14 10:10:07 -07:00
access_group.py feat(access-groups): grant Claude Code skills through access groups (#46294) 2026-10-14 23:18:50 +00:00
adapter.py refactor(types): import CustomLogger and CustomLLM under TYPE_CHECKING in adapter TypedDicts (#45774) 2026-10-10 19:07:11 +00:00
agents.py refactor: expose public hidden_params accessors (#44668) 2026-10-07 00:01:33 -07:00
caching.py fix(caching): never store or serve a chat completion with no choices (#44709) 2026-10-06 20:07:30 +00:00
completion.py refactor: expose core private helpers under public names (#44871) 2026-10-07 00:05:27 +00:00
compression.py
decisions.py feat(typesafe): support Jev 1.14 images, box questions, and long-context headers (#46596) 2026-10-14 22:29:48 -07:00
embedding.py perf(types): defer pydantic schema builds via shared LiteLLMBaseModel (#44720) 2026-10-05 22:10:08 -07:00
files.py
fine_tuning.py
guardrail_base_init.py refactor(types): replace Any with real types across 54 more backend files 2026-08-29 19:00:43 +00:00
guardrails.py feat(guardrails): add Airia guardrail integration (#40784) 2026-10-14 15:45:47 -07:00
lar1.py perf(types): defer pydantic schema builds via shared LiteLLMBaseModel (#44720) 2026-10-05 22:10:08 -07:00
litellm_params.py fix(websearch_interception): drop stream_options when forcing stream=False and replay it on the final stream (#45433) 2026-10-09 22:16:02 -07:00
mcp.py feat(mcp): separate confidential client for ID-JAG leg 2 (#46211) 2026-10-14 10:04:31 -04:00
memory_management.py perf(types): defer pydantic schema builds via shared LiteLLMBaseModel (#44720) 2026-10-05 22:10:08 -07:00
model_insights.py perf(types): defer pydantic schema builds via shared LiteLLMBaseModel (#44720) 2026-10-05 22:10:08 -07:00
object_permission.py feat(proxy): granular key/team access control for Claude Code marketplace plugins (#40518) 2026-09-10 14:26:09 -07:00
policy_engine.py
rag.py perf(types): defer pydantic schema builds via shared LiteLLMBaseModel (#44720) 2026-10-05 22:10:08 -07:00
realtime.py perf(types): defer pydantic schema builds via shared LiteLLMBaseModel (#44720) 2026-10-05 22:10:08 -07:00
rerank.py refactor: expose public hidden_params accessors (#44668) 2026-10-07 00:01:33 -07:00
roi_calculator.py perf(types): defer pydantic schema builds via shared LiteLLMBaseModel (#44720) 2026-10-05 22:10:08 -07:00
roi_observed.py perf(types): defer pydantic schema builds via shared LiteLLMBaseModel (#44720) 2026-10-05 22:10:08 -07:00
router.py feat(xai): bill grok voice realtime sessions (#46600) 2026-10-15 05:27:02 +00:00
router_weights.py fix(router): honor team and key provider weights 2026-09-14 23:31:52 -07:00
scheduler.py
search.py fix(cost): price AgentCore web search at $7 per 1,000 queries (#46571) 2026-10-14 18:30:39 -07:00
services.py perf(types): defer pydantic schema builds via shared LiteLLMBaseModel (#44720) 2026-10-05 22:10:08 -07:00
tag_management.py perf(types): defer pydantic schema builds via shared LiteLLMBaseModel (#44720) 2026-10-05 22:10:08 -07:00
tool_management.py perf(types): defer pydantic schema builds via shared LiteLLMBaseModel (#44720) 2026-10-05 22:10:08 -07:00
utils.py feat(typesafe): support Jev 1.14 images, box questions, and long-context headers (#46596) 2026-10-14 22:29:48 -07:00
vector_store_files.py
vector_stores.py perf(types): defer pydantic schema builds via shared LiteLLMBaseModel (#44720) 2026-10-05 22:10:08 -07:00
workload_identity.py feat(anthropic): workload identity federation and pluggable identity sources (#44448) 2026-10-03 17:08:30 -07:00