HoWToBench - Poem: 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 poem 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 poem writing instructions of HoWToBench; higher is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 10 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1GPT-4o (2024-11-20)6.34#859
2HoWToBench DeepSeek-R1 (checkpoint unspecified)6
3O3 Mini (2025-01-31)5.81#496
4HoWToBench DeepSeek-V3 (checkpoint unspecified)5.6#666
5GLM-4-Plus-2501115.47#831
6Gemini 2.0 Flash5.42#723
7Claude 3.5 Sonnet (20241022)5.41#648
8Doubao-pro-2412255.15#810
9Claude 3 Haiku (20240307)4.58#2396
10Llama 3.3 70B Instruct4.2#1385

No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=howtobench-poem · How It Works · Data refreshed daily, snapshot 2026-11-21.