reasoning-benchmark

Open repos and models their makers tagged reasoning-benchmark, most starred first. The same for everyone.

GitHubRepo

FP-AMB

First-Person Agent Memory Bench. 10 Categories including fact recall, multi-hop links, temporal reasoning, fact overwrites, speaker traps, refusal, credibility, and agentic tool usage. 540K token / 60 session corpus, all in first person. Dynamic output-answer-key portion. Comprehensive report with visuals and miss breakdown.

  • 11
  • Python
  • MIT
  • 3 mo ago
munch2u-a11yTake it
Search GitHub and Hugging Face for reasoning-benchmark

Get , open it once and sign in. Then come back here and press the button again.