RPCBench - Domain - Yelp Local - Fact Fabrication: leaderboard
Metric: Fact Fabrication Rate (%). Source: github.com. Saturation forecast: Estimated already saturated. 11 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 3.5 Plus | 1.15 |
| 2 | Qwen 3.5 397B A17B | 1.3 |
| 3 | Qwen 3.5 35B A3B | 1.37 |
| 4 | Qwen 3.5 122B A10B | 2.21 |
| 5 | DeepSeek V4 Flash | 4.58 |
| 6 | GPT-5.5 | 5.11 |
| 7 | DeepSeek V4 Pro | 5.5 |
| 8 | Gemini 3.1 Pro (Preview) | 6.72 |
| 9 | Claude Sonnet 4.6 | 9.69 |
| 10 | Llama 3.1 70B Instruct | 53.97 |
| 11 | Llama 3.1 8B Instruct | 62.37 |
Interactive version: theaggregate.ai/benchmark?slug=rpcbench-domain-yelp-local-fact-fabrication · How It Works · Data refreshed daily, snapshot 2026-09-24.