mirror of
https://github.com/ggozad/oterm.git
synced 2026-10-13 18:53:21 +02:00
pydantic-ai defaults allow_sampling=True with sampling_model=None, so a server that issues a sampling request crashes inside the callback. Until oterm decides how to pick a sampling model, advertise sampling as off so the protocol rejects requests cleanly, and document the gap. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> |
||
|---|---|---|
| .. | ||
| index.md | ||