OmniGAIA - History: leaderboard
The "History" subtask of OmniGAIA. Omni-modal general AI assistant benchmark for long-horizon agents using vision, audio, language, web search, browser, and code-execution tasks.
Metric: History Accuracy (%). Source: huggingface.co. Saturation forecast: Around January 2027. 16 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3 Pro | 62.1 | #80 |
| 2 | Gemini 3 Flash | 44.8 | #99 |
| 3 | Gemini 2.5 Pro | 32.8 | #152 |
| 4 | Gemini 2.5 Flash Lite | 14.9 | #422 |
| 5 | Qwen2.5-Omni-7B | 7.5 | #624 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=omnigaia-history · How It Works · Data refreshed daily, snapshot 2026-10-20.