HoWToBench DeepSeek-V3 (checkpoint unspecified): benchmark results
Provider: Other.
Aggregate Rating 1641 ± 38, rank #685 of 10916 rated models, from 15 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| HoWToBench - Letters | 6.18 | Tree-of-Writing score: GPT-4o rates each content trait 1-10 | 88.9 |
| HoWToBench - Contract | 5.33 | Tree-of-Writing score: GPT-4o rates each content trait 1-10 | 77.8 |
| HoWToBench - Plan | 5.11 | Tree-of-Writing score: GPT-4o rates each content trait 1-10 | 66.7 |
| HoWToBench - Poem | 5.6 | Tree-of-Writing score: GPT-4o rates each content trait 1-10 | 66.7 |
| HoWToBench - Report | 5.55 | Tree-of-Writing score: GPT-4o rates each content trait 1-10 | 66.7 |
| HoWToBench - Speech | 5.74 | Tree-of-Writing score: GPT-4o rates each content trait 1-10 | 66.7 |
| HoWToBench - Comment | 5.53 | Tree-of-Writing score: GPT-4o rates each content trait 1-10 | 55.6 |
| HoWToBench - Completion | 5.44 | Tree-of-Writing score: GPT-4o rates each content trait 1-10 | 44.4 |
| HoWToBench - Guided Writing | 5.52 | Tree-of-Writing score: GPT-4o rates each content trait 1-10 | 44.4 |
| HoWToBench - Open Writing | 5.31 | Tree-of-Writing score: GPT-4o rates each content trait 1-10 | 44.4 |
| HoWToBench - Prose | 5.16 | Tree-of-Writing score: GPT-4o rates each content trait 1-10 | 44.4 |
| HoWToBench - Regulation | 4.87 | Tree-of-Writing score: GPT-4o rates each content trait 1-10 | 44.4 |
Interactive version: theaggregate.ai/model?slug=howtobench-deepseek-v3-checkpoint-unspecified · How It Works · Data refreshed daily, snapshot 2026-11-08.