RPCBench - Error Group - Inconsistent Premise - Composite Premise Critique: leaderboard
Metric: Composite Score (0-100). Source: github.com. Saturation forecast: Around December 2026. 11 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | DeepSeek V4 Pro | 69.63 |
| 2 | Claude Sonnet 4.6 | 68.73 |
| 3 | Qwen 3.5 397B A17B | 68.44 |
| 4 | Qwen 3.5 Plus | 67.95 |
| 5 | Qwen 3.5 122B A10B | 67.65 |
| 6 | Gemini 3.1 Pro (Preview) | 66.78 |
| 7 | DeepSeek V4 Flash | 62.99 |
| 8 | Qwen 3.5 35B A3B | 61.9 |
| 9 | GPT-5.5 | 61.51 |
| 10 | Llama 3.1 70B Instruct | 14.04 |
| 11 | Llama 3.1 8B Instruct | 10.03 |
Interactive version: theaggregate.ai/benchmark?slug=rpcbench-error-group-inconsistent-premise-composite-premise-critique · How It Works · Data refreshed daily, snapshot 2026-09-24.