Recall
Unstoppable intelligence network for AI agents to store, share, and trade knowledge.
Pinned Loading
Repositories
Showing 10 of 67 repositories
- llm-tournament-benchmark Public
Pairwise LLM benchmark, August 2025: 45 scenarios across 8 skills, 51 models, Swiss tournaments, LLM judges and OpenSkill ratings. Results in this repo are quick runs of 4 models. To rerun 2 summarization scenarios, download arXiv 2412.14498 and 2504.17028 into documents/. Their licenses do not let this repo include them.
- electric-pelican Public archive
A platform for evaluating and intelligently routing AI models using real head-to-head performance data and hierarchical capability scores
- air-coding-arena Public archive
Reference application tracking Linear coding-agent competitions, GitHub pull requests, and OpenSkill leaderboards.
- astrochicken Public archive
Reference platform for orchestrating coding-agent workflows and local software factories
- skills-are-all-you-need Public archive
- survival-receipts Public archive