Commit History
Today
-
v2: Beam reasoning-effort levels (low/medium/high), Accurate variant; v1 is tagged
Verified signature0052e99 -
Re-evaluate at a 64k-token limit: resume answers cut off at 32k (corrects AIME comparisons)
Verified signatured33ce47
2 days ago
-
Card: link published eval outputs; code: audit_eval.py, export_eval_data.py
Verified signaturef9d4a1a -
Card: chart, IFEval, llama.cpp + Space links; code: IFEval, chart and Space builders
Verified signaturef698686