List your company

Mathematical Reasoning on FormInv N=103 theorems v1

96.3AccuracyDeepSeek V3
85.27688.1389193.862May 27, 2026

Evaluation results

MethodLinks
DeepSeek V3Coverage=≥98.6%2026.0596.383.51.5
o1-miniCoverage=≥98.6%2026.0595.279.60.4
Gemini 1.5F.Coverage=96.4%2026.0594.282.52.5
Sonnet 3.5Coverage=≥98.6%2026.059478.64.6
GPT-4oCoverage=≥98.6%2026.0593.179.62.4
GPT-4o-miniCoverage=≥98.6%2026.0592.673.84.2
DeepSeek R1Coverage=≥98.6%2026.0591.975.72.3
Llama 70BCoverage=≥98.6%2026.0589.8671
Haiku 3.5Coverage=≥98.6%2026.0585.754.44.4