jnjpng e3eafb1977 fix: re-raise LLMError before wrapping with handle_llm_error (#9482) ...

LLMError exceptions are already properly formatted errors that should
propagate directly. Without this check, they get unnecessarily wrapped
by handle_llm_error, losing their original error information.

2026-02-24 10:52:07 -08:00

..

letta_llm_adapter.py

feat: add log probabilities from OpenAI-compatible servers and SGLang native endpoint (#9240)

2026-02-24 10:52:07 -08:00

letta_llm_request_adapter.py

feat: add log probabilities from OpenAI-compatible servers and SGLang native endpoint (#9240)

2026-02-24 10:52:07 -08:00

letta_llm_stream_adapter.py

fix: catch contextwindowexceeded error on gemini (#9450)

2026-02-24 10:52:07 -08:00

sglang_native_adapter.py

feat: add log probabilities from OpenAI-compatible servers and SGLang native endpoint (#9240)

2026-02-24 10:52:07 -08:00

simple_llm_request_adapter.py

fix: re-raise LLMError before wrapping with handle_llm_error (#9482)

2026-02-24 10:52:07 -08:00

simple_llm_stream_adapter.py

fix: re-raise LLMError before wrapping with handle_llm_error (#9482)

2026-02-24 10:52:07 -08:00