Agent2 RL-Bench - GSM8K: leaderboard
Agentic post-training engineering on Agent^2 RL-Bench (2026): a coding agent driven by the model has 12 hours in an isolated GPU workspace to post-train Qwen3-8B-Base, and the grading server keeps the best checkpoint score of the run. GSM8K test accuracy of the trained model, in percent; higher is better.
Metric: GSM8K test accuracy (%) of the Qwen3-8B-Base model after the coding agent (driver LLM in a CLI scaffold) post-trains it for 12 hours in an isolated GPU workspace, best score the grading server returned over the run, single run; higher is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 6 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.2 | 89.46 | #185 |
| 2 | Claude Opus 4.6 | 88.78 | #113 |
| 3 | Claude Sonnet 4.5 | 88.7 | #261 |
| 4 | GPT-5.4 | 84 | #135 |
| 5 | GPT-4o | 83.32 | #751 |
| 6 | Gemini 2.5 Flash | 70.13 | #463 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=agent2-rl-bench-gsm8k · How It Works · Data refreshed daily, snapshot 2026-11-21.