RPCBench - Domain - Goodreads Dual-Domain - Conditional Localization: leaderboard
Metric: Localization Accuracy (%). Source: github.com. Saturation forecast: Estimated already saturated. 11 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Sonnet 4.6 | 90.74 |
| 2 | Qwen 3.5 397B A17B | 89.92 |
| 3 | GPT-5.5 | 89.29 |
| 4 | Qwen 3.5 Plus | 88.64 |
| 5 | Gemini 3.1 Pro (Preview) | 88.08 |
| 6 | Qwen 3.5 122B A10B | 87.77 |
| 7 | DeepSeek V4 Flash | 87.14 |
| 8 | Qwen 3.5 35B A3B | 86.02 |
| 9 | DeepSeek V4 Pro | 85.06 |
| 10 | Llama 3.1 8B Instruct | 52.06 |
| 11 | Llama 3.1 70B Instruct | 49.32 |
Interactive version: theaggregate.ai/benchmark?slug=rpcbench-domain-goodreads-dual-domain-conditional-localization · How It Works · Data refreshed daily, snapshot 2026-09-24.