LiveBench Simplify: leaderboard
The model is given a recent article from The Guardian and asked to simplify it while obeying randomly drawn constraints such as a word count, required keywords or banned words. This is the simplify task in LiveBench's instruction following category, and the score counts only adherence to those instructions, checked without a model judge. Scores run 0 to 100 and higher is better.
Metric: Score. Source: livebench.ai. Saturation forecast: Around 2031. 71 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.6 Flash (High) | 75.7 | #51 (Gemini 3.6 Flash) |
| 2 | Gemini 3.8 Flash (High) | 75.4 | #21 (Gemini 3.8 Flash) |
| 3 | Gemini 3.7 Flash (High) | 74.62 | #23 (Gemini 3.7 Flash) |
| 4 | Muse Spark 1.3 (xHigh) | 72.52 | #24 (Muse Spark 1.3) |
| 5 | Claude Sonnet 5.5 (xHigh) | 72.37 | #11 (Claude Sonnet 5.5) |
| 6 | Nemotron 3 Ultra | 72.28 | #235 |
| 7 | Claude Fable 5 (Max) | 72.03 | #16 (Claude Fable 5) |
| 8 | GPT-6.1 Sol (Max) | 71.97 | #7 (GPT-6.1 Sol) |
| 9 | Gemini 3.1 Pro (Preview) (High) | 71.05 | #58 (Gemini 3.1 Pro (Preview)) |
| 10 | GPT-6 Astra (Max) | 70.53 | #6 (GPT-6 Astra) |
| 11 | GPT-5.6 Sol (Max) | 70.52 | #19 (GPT-5.6 Sol) |
| 12 | Qwen3.8-Flash-Next | 70.25 | #72 |
| 13 | Grok 4.5 | 70.17 | #48 |
| 14 | Claude Fable 5.1 (Max) | 70.17 | #9 (Claude Fable 5.1) |
| 15 | GPT-5.4 (xHigh) | 70 | #79 (GPT-5.4) |
No result here: #8 Gemini 4 Argon, #10 Muse Wonder Full.
Interactive version: theaggregate.ai/benchmark?slug=livebench-simplify · How It Works · Data refreshed daily, snapshot 2026-10-20.