litellm/tests/integration/messages_endpoint/chat_bridge
Nate Armstrong 23025c92ba
fix(anthropic): emit one tool_use block per call when a /v1/messages stream chunk carries several (#46492)
Providers like Mistral put every parallel tool call, with full arguments, into one
chat-completions chunk. The /v1/messages adapter opened one tool_use block for the
first call and concatenated every call's arguments into it, so clients got invalid
JSON like {...}{...}. _CombinedChunkSplitter now splits such a chunk into one chunk
per call. The first call streams live and calls opened behind it are held, gathering
their fragments by index, until a non-fragment chunk arrives or the stream ends

Fixes #44029
2026-10-15 04:27:51 +00:00
..
test_anthropic_messages_fireworks_stop_wire.py test(integration): group /v1/messages contracts under tests/integration/messages_endpoint (#43352) 2026-09-26 16:00:32 -07:00
test_anthropic_messages_hosted_vllm_reasoning_wire.py fix(hosted_vllm): keep reasoning_content on replayed assistant messages (#43599) 2026-09-30 13:12:28 -07:00
test_chat_bridge_mid_stream_failure_bookkeeping_wire.py fix(anthropic): let /v1/messages mid-stream failures reach the proxy failure boundary (#44800) 2026-10-07 17:42:35 -07:00
test_chat_bridge_multi_tool_call_blocks_wire.py fix(anthropic): emit one tool_use block per call when a /v1/messages stream chunk carries several (#46492) 2026-10-15 04:27:51 +00:00
test_chat_bridge_pre_content_retry_wire.py fix(router): retry a /v1/messages stream the provider drops before the first content chunk (#44276) 2026-10-05 22:53:39 +00:00