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...

Fibble3 Arena — leaderboard

Metric: Win Rate (%). Source: drchangliu.github.io. 43 models tracked.

Top models

#ModelScore
1DeepSeek R133.33
2Gemini 3.1 Pro (Preview)25
3Kimi K2.58.57
4Qwen 3.5 27B3.12
5Llama 3.1 8B2.86
6Gemini 2.5 Pro0
7GPT-4o0
8GPT-4o Mini0
9GPT-50
10GPT-5 Mini0
11Gemma 3 27B0
12Qwen 3 8B0
13Gemma 3 4B0
14Qwen 3 30B A3B0
15Qwen 3 14B0

Interactive version: theaggregate.ai/benchmark?slug=fibble3-arena · 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.