riddle: switch to abliterated model to stop shadow/fire loops
Lint / Shell (shellcheck) (push) Successful in 11s
Lint / JS (eslint) (push) Successful in 7s
Lint / Python (ruff) (push) Successful in 5s
Lint / Python deps (pip-audit) (push) Successful in 43s
Lint / Secret scan (gitleaks) (push) Successful in 5s

phi4-mini is too conservative and defaults to the same 2-3 answers.
Use BALL_MODEL (abliterated Llama 3.2) like WYR does.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
This commit is contained in:
2026-04-22 21:48:32 -04:00
parent 63f1bfda49
commit 8bbcc0530f
+1 -1
View File
@@ -1806,7 +1806,7 @@ async def _generate_riddle() -> dict | None:
async with session.post( async with session.post(
f"{OLLAMA_URL}/api/chat", f"{OLLAMA_URL}/api/chat",
json={ json={
"model": ASK_MODEL, "model": BALL_MODEL,
"stream": False, "stream": False,
"messages": [ "messages": [
{"role": "system", "content": system_msg}, {"role": "system", "content": system_msg},