TeamGym - EvidenceBench - Single-Agent Full-Context: leaderboard
TeamGym measures how agents combine information to solve tasks. Results distinguish individual agents, cooperating agents and how much information each receives. Episode success (%).
Metric: Episode success (%). Source: arxiv.org. Saturation forecast: Around May 2027. 8 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.5 | 31.1 | #32 |
| 2 | Claude Sonnet 4.5 | 24.4 | #146 |
| 3 | Claude Haiku 4.5 | 20 | #283 |
| 4 | Gemma 3 27B (IT) | 8.9 | #521 |
| 5 | Qwen 3 32B | 8.9 | #443 |
| 6 | Qwen 3 235B A22B | 8.9 | #317 |
| 7 | Gemma 3 12B (IT) | 0 | #662 |
| 8 | Gemma 3 4B (IT) | 0 | #1010 |
No result here: #3 Claude Opus 5.5, #5 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=teamgym-evidencebench-single-agent-full-context · How It Works · Data refreshed daily, snapshot 2026-10-29.