RPCBench - Domain - MovieLens-1M - Evidence Distortion: leaderboard

Metric: Evidence Distortion Rate (%). Source: github.com. Saturation forecast: Around 2029. 11 models tracked.

Top models

#ModelScore
1Llama 3.1 8B Instruct23.59
2Llama 3.1 70B Instruct32.47
3Gemini 3.1 Pro (Preview)33.55
4Qwen 3.5 Plus34.42
5Qwen 3.5 35B A3B34.63
6Qwen 3.5 397B A17B35.06
7Qwen 3.5 122B A10B35.06
8GPT-5.536.36
9DeepSeek V4 Pro38.96
10Claude Sonnet 4.640.26
11DeepSeek V4 Flash40.91

Interactive version: theaggregate.ai/benchmark?slug=rpcbench-domain-movielens-1m-evidence-distortion · How It Works · Data refreshed daily, snapshot 2026-09-24.