JudgeSense - Factuality: leaderboard
JudgeSense (2026) measures how often an LLM judge gives the same verdict when one judging request is worded two equivalent ways, at temperature 0, on items from human-labelled corpora. Agreement across the two wordings when judging TruthfulQA statements as accurate or inaccurate, from 0 to 1; higher is better.
Metric: Judge Sensitivity Score: share of items on which the judge returns the same verdict under two reworded instruction templates that ask for the same judgement, over the pairs it answered (refusals and malformed outputs left out); temperature 0 (claude-opus-4-7 at the provider default), 1,024 output tokens, on 250 TruthfulQA statements judged accurate or inaccurate; higher is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 25 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemma 4 31B (IT) | 0.98 | #350 |
| 2 | Gemini 3.7 Flash | 0.976 | #41 |
| 3 | Claude Opus 4.7 | 0.96 | #85 |
| 4 | Qwen 3.7 Flash (2026-07-15) | 0.96 | |
| 5 | Claude Haiku 4.5 (20251001) | 0.956 | #543 |
| 6 | Claude Sonnet 4.5 | 0.952 | #261 |
| 7 | Llama 3.3 70B Instruct | 0.952 | #1385 |
| 8 | Qwen 3 32B | 0.952 | #972 |
| 9 | Kimi K3 | 0.948 | #61 |
| 10 | Qwen 3.6 35B A3B | 0.932 | #481 |
| 11 | Llama 4 Maverick Instruct FP8 | 0.916 | #1023 |
| 12 | Qwen 2.5 72B Instruct | 0.908 | #1117 |
| 13 | DeepSeek V4 Pro (0813) | 0.904 | #97 |
| 14 | Gemini 2.5 Flash | 0.896 | #463 |
| 15 | Llama 4 Scout Instruct | 0.892 | #1579 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=judgesense-factuality · How It Works · Data refreshed daily, snapshot 2026-11-21.