Skip to content

The Aggregate

Unified LLM rankings, daily

Models What's New Trends Benchmarks Skill Maps
How It Works Guesswork Metrics
The Aggregate
Loading data...

UGI - Writing — leaderboard

Metric: Writing Score. Source: huggingface.co. 1234 models tracked.

Top models

#ModelScore
1Gemini 3.5 Flash (Minimal)72.47
2Claude Opus 4.6 (Adaptive Reasoning, Max Effort)70.7
3Claude Opus 4.5 (20251101) (Thinking)70.34
4Claude Opus 4.1 (Non-reasoning)69.91
5Gemini 3.5 Flash (High)69.71
6GPT-5.5 (xHigh)69.66
7GPT-5.4 (High)68.82
8GPT-5.4 (xHigh)68.65
9DeepSeek V4 Pro (Reasoning)68.4
10Claude Opus 4 (Non-reasoning)68.24
11Gemini 3.1 Pro (Preview) (High)68.22
12Claude 3.7 Sonnet (20250219) (Thinking)68.01
13GPT-5.5 (High)67.91
14Gemini 3 Flash (Preview) (Medium)67.71
15GPT-4.167.48

Interactive version: theaggregate.ai/benchmark?slug=ugi-writing · How the rankings work · Data refreshed daily, snapshot 2026-07-22.

Built by Mikhail Doroshenko — AI researcher, co-author of Humanity’s Last Exam. Independent project; no affiliation with any AI lab.