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...

GRACE — leaderboard

Metric: Avg Step (self-reported). Source: benchmarklist.com. 13 models tracked.

Top models

#ModelScore
1Gemini 3.1 Pro (Preview)81.46
2GPT-5.478.53
3Gemma 4 31B77.46
4Qwen 3.5 35B A3B75.7
5Gemini 3 Flash (Preview)75.19
6Qwen 3.5 27B74.59
7GPT-5.4 Mini66.85
8Qwen 3 14B56.6
9Qwen 3 8B49.56
10Llama 3.1 8B45.76

Interactive version: theaggregate.ai/benchmark?slug=grace · 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.