WritingBench: leaderboard
Generative writing benchmark covering writing quality, style control, factuality, and instruction following across writing tasks.
Metric: Score (%). Source: writingbench-writingbench.hf.space. Source: llm-stats.com. Saturation forecast: Estimated already saturated. 16 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3 235B A22B 2507 (Thinking) | 88.3 | #260 (Qwen 3 235B A22B 2507) |
| 2 | Qwen 3 Next 80B A3B Instruct | 87.3 | #368 |
| 3 | Qwen 3 VL 235B A22B (Thinking) | 86.7 | #238 (Qwen 3 VL 235B A22B) |
| 4 | Qwen 3 VL 32B (Thinking) | 86.2 | #294 (Qwen 3 VL 32B) |
| 5 | Qwen 3 VL 235B A22B Instruct | 85.5 | #273 |
| 6 | Qwen 3 VL 8B (Thinking) | 85.5 | |
| 7 | Qwen 3 235B A22B 2507 Instruct | 85.2 | #300 |
| 8 | Qwen 3 VL 30B A3B (Thinking) | 85.2 | #351 (Qwen 3 VL 30B A3B) |
| 9 | Qwen 3 Next 80B A3B (Thinking) | 84.6 | #320 (Qwen 3 Next 80B A3B) |
| 10 | Qwen 3 VL 4B (Thinking) | 84 | #480 (Qwen 3 VL 4B) |
| 11 | Qwen 3 VL 8B Instruct | 83.1 | #420 |
| 12 | Qwen 3 VL 32B Instruct | 82.9 | #283 |
| 13 | Qwen 3 VL 30B A3B Instruct | 82.6 | #384 |
| 14 | Qwen 3 VL 4B Instruct | 82.5 | #515 |
| 15 | Qwen 3 14B | 78 | #537 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=writingbench · How It Works · Data refreshed daily, snapshot 2026-10-20.