LiveBench Paraphrase: leaderboard
The model is given a recent article from The Guardian and asked to paraphrase it while obeying randomly drawn constraints such as a word limit or elements it must include. It is a single-turn task with the article supplied in context, and scoring is an automatic check of whether the constraints were met rather than a judgement of writing quality. Higher is better.
Metric: Score. Source: livebench.ai. Saturation forecast: Not forecast. 71 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.8 Flash (High) | 80 | #21 (Gemini 3.8 Flash) |
| 2 | Gemini 3.1 Pro (Preview) (High) | 79.87 | #58 (Gemini 3.1 Pro (Preview)) |
| 3 | Qwen3.8-Flash-Next | 79.47 | #72 |
| 4 | Muse Spark 1.3 (xHigh) | 78.62 | #24 (Muse Spark 1.3) |
| 5 | GPT-6 Astra (Max) | 78.25 | #6 (GPT-6 Astra) |
| 6 | Gemini 3.7 Flash (High) | 78.07 | #23 (Gemini 3.7 Flash) |
| 7 | Grok 4.7 (xHigh) | 78.05 | #25 (Grok 4.7) |
| 8 | Claude Fable 5 (Max) | 77.65 | #16 (Claude Fable 5) |
| 9 | Nemotron 3 Ultra | 77.38 | #235 |
| 10 | Gemini 3.5 Flash (High) | 75.52 | #60 (Gemini 3.5 Flash) |
| 11 | Qwen 3.8 27B | 74.9 | #105 |
| 12 | Kimi K3 | 74.37 | #30 |
| 13 | Gemini 3.6 Flash (High) | 74.32 | #51 (Gemini 3.6 Flash) |
| 14 | Muse Spark 1.2 (xHigh) | 74.3 | #62 (Muse Spark 1.2) |
| 15 | Muse Wonder (Max) | 74.07 | #15 (Muse Wonder) |
No result here: #8 Gemini 4 Argon, #10 Muse Wonder Full.
Interactive version: theaggregate.ai/benchmark?slug=livebench-paraphrase · How It Works · Data refreshed daily, snapshot 2026-10-20.