Repository navigation

Qwen3 thinking in the agent loop; server-like Sonos eval fixture and scoring - #2

Merged

ilnmtlbnm merged 12 commits into

mainfrom

Oct 18, 2026

Merged

Qwen3 thinking in the agent loop; server-like Sonos eval fixture and scoring#2
ilnmtlbnm merged 12 commits into
mainfrom
qwen3-thinking

Commits on Oct 15, 2026
Commits on Oct 17, 2026
- agent eval: a mutating call rejected with an error result does not count as a mutation
  ilnmtlbnm
  committed
  c8c6e63
Commits on Oct 18, 2026
- docs: Sonos runs with rejected mutating calls not counted as mutations (Qwen3-1.7B, thinking on, default, clear, clear cap 512)
  ilnmtlbnm
  committed
  5961c11