HoWToBench - Comment: leaderboard
HoWToBench (2026) tests long-form Chinese writing against curated human references; GPT-4o rates content traits 1 to 10 and format traits 0, 5 or 10, and the Tree-of-Writing method combines them with explicit per-instruction weights. Mean score on the comment genre instructions; higher is better.
Metric: Tree-of-Writing score: GPT-4o rates each content trait 1-10 against the human reference and format traits 0, 5 or 10, combined with per-instruction LLM-assigned weights that sum to 1, mean over the comment writing instructions of HoWToBench; higher is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 10 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | O3 Mini (2025-01-31) | 6.02 | #502 |
| 2 | GPT-4o (2024-11-20) | 5.98 | #860 |
| 3 | HoWToBench DeepSeek-R1 (checkpoint unspecified) | 5.95 | |
| 4 | Claude 3.5 Sonnet (20241022) | 5.54 | #651 |
| 5 | HoWToBench DeepSeek-V3 (checkpoint unspecified) | 5.53 | #669 |
| 6 | Gemini 2.0 Flash | 5.36 | #731 |
| 7 | GLM-4-Plus-250111 | 5.36 | #836 |
| 8 | Doubao-pro-241225 | 5.3 | #817 |
| 9 | Claude 3 Haiku (20240307) | 5.1 | #2415 |
| 10 | Llama 3.3 70B Instruct | 4.65 | #1394 |
No result here: #2 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=howtobench-comment · How It Works · Data refreshed daily, snapshot 2026-11-22.