mirror of
https://github.com/BerriAI/litellm.git
synced 2026-10-04 02:31:27 +00:00
- Add VertexBase.invalidate_credentials; remove (creds, project_id) and (creds, None) when project_id is set (#23512). - async_anthropic_messages_handler: on httpx 401 for vertex_ai, invalidate before _handle_error so retries can load_auth/refresh. - Use module-level httpx in invalidation helper (no redundant inline import). - Unit tests in test_vertex_llm_base.py and test_llm_http_handler.py (mocked). Fixes #23512 Made-with: Cursor |
||
|---|---|---|
| .. | ||
| test_aiohttp_cleanup_closed.py | ||
| test_aiohttp_handler.py | ||
| test_aiohttp_transport.py | ||
| test_gemini_session_leak.py | ||
| test_http_handler.py | ||
| test_llm_http_handler.py | ||
| test_mock_transport.py | ||