RPCBench - Error Group - Unverifiable Premise - Conditional Strategy Quality: leaderboard
Metric: Strategy Quality (0-100). Source: github.com. Saturation forecast: Around December 2026. 11 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | DeepSeek V4 Pro | 84.5 |
| 2 | DeepSeek V4 Flash | 82.41 |
| 3 | Qwen 3.5 Plus | 80.49 |
| 4 | Qwen 3.5 397B A17B | 79.87 |
| 5 | Qwen 3.5 122B A10B | 79.18 |
| 6 | Claude Sonnet 4.6 | 78.73 |
| 7 | GPT-5.5 | 77.17 |
| 8 | Qwen 3.5 35B A3B | 75.09 |
| 9 | Gemini 3.1 Pro (Preview) | 67.12 |
| 10 | Llama 3.1 70B Instruct | 51.18 |
| 11 | Llama 3.1 8B Instruct | 43.96 |
Interactive version: theaggregate.ai/benchmark?slug=rpcbench-error-group-unverifiable-premise-conditional-strategy-quality · How It Works · Data refreshed daily, snapshot 2026-09-24.