teahubxyzandClaude 3086cb52ff labs/aicharts: fast non-thinking model for specs (#39) ...

Live: gemma-4 took 70s+ and returned empty content on harder prompts (budget spent
thinking). Llama 3.3 70B fp8-fast answers the spec directly; still overridable.


Claude-Session: https://claude.ai/code/session_01VDkN7ukK1kXkqov8FaZxva

Co-authored-by: Claude <noreply@anthropic.com>

2026-10-13 14:01:33 +00:00

..

ISSUE_TEMPLATE

Venture loop: launch products, not pages

2026-10-11 15:48:21 -07:00

time

labs/aicharts: fast non-thinking model for specs (#39)

2026-10-13 14:01:33 +00:00

workflows

labs/aicharts: fix model spelling drift, show the model's answer on failure (#37)

2026-10-13 13:47:49 +00:00