llama-stack-mirror/llama_stack/providers/inline
ehhuang 14a94e9894
Some checks failed
Integration Auth Tests / test-matrix (oauth2_token) (push) Failing after 2s
SqlStore Integration Tests / test-postgres (3.12) (push) Failing after 1s
Python Package Build Test / build (3.12) (push) Failing after 2s
Integration Tests (Replay) / Integration Tests (, , , client=, ) (push) Failing after 5s
Test External Providers Installed via Module / test-external-providers-from-module (venv) (push) Has been skipped
API Conformance Tests / check-schema-compatibility (push) Successful in 10s
Vector IO Integration Tests / test-matrix (push) Failing after 5s
Python Package Build Test / build (3.13) (push) Failing after 3s
SqlStore Integration Tests / test-postgres (3.13) (push) Failing after 9s
Test External API and Providers / test-external (venv) (push) Failing after 6s
Unit Tests / unit-tests (3.12) (push) Failing after 5s
Unit Tests / unit-tests (3.13) (push) Failing after 6s
UI Tests / ui-tests (22) (push) Successful in 33s
Pre-commit / pre-commit (push) Successful in 1m27s
fix: responses <> chat completion input conversion (#3645)
# What does this PR do?

closes #3268
closes #3498

When resuming from previous response ID, currently we attempt to convert
from the stored responses input to chat completion messages, which is
not always possible, e.g. for tool calls where some data is lost once
converted from chat completion message to repsonses input format.

This PR stores the chat completion messages that correspond to the
_last_ call to chat completion, which is sufficient to be resumed from
in the next responses API call, where we load these saved messages and
skip conversion entirely.

Separate issue to optimize storage:
https://github.com/llamastack/llama-stack/issues/3646

## Test Plan
existing CI tests
2025-10-02 16:01:08 -07:00
..
agents fix: responses <> chat completion input conversion (#3645) 2025-10-02 16:01:08 -07:00
batches feat(batches, completions): add /v1/completions support to /v1/batches (#3309) 2025-09-05 11:59:57 -07:00
datasetio chore(misc): make tests and starter faster (#3042) 2025-08-05 14:55:05 -07:00
eval feat: update eval runner to use openai endpoints (#3588) 2025-09-29 13:13:53 -07:00
files/localfs fix(expires_after): make sure multipart/form-data is properly parsed (#3612) 2025-09-30 16:14:03 -04:00
inference chore: remove /v1/inference/completion and implementations (#3622) 2025-10-01 11:36:53 -04:00
ios/inference feat(tools)!: substantial clean up of "Tool" related datatypes (#3627) 2025-10-02 15:12:03 -07:00
post_training chore(pre-commit): add pre-commit hook to enforce llama_stack logger usage (#3061) 2025-08-20 07:15:35 -04:00
safety feat: use /v1/chat/completions for safety model inference (#3591) 2025-09-30 11:01:44 -07:00
scoring chore: use openai_chat_completion for llm as a judge scoring (#3635) 2025-10-01 09:44:31 -04:00
telemetry chore: Remove debug logging from telemetry adapter (#3643) 2025-10-01 15:16:23 -07:00
tool_runtime feat(tools)!: substantial clean up of "Tool" related datatypes (#3627) 2025-10-02 15:12:03 -07:00
vector_io refactor: use generic WeightedInMemoryAggregator for hybrid search in SQLiteVecIndex (#3303) 2025-09-02 10:38:35 -07:00
__init__.py impls -> inline, adapters -> remote (#381) 2024-11-06 14:54:05 -08:00