Sandboxed Coding Agents are Competitive Omni-m: leaderboard
Metric: OmniGAIA Avg. (self-reported). Source: benchmarklist.com. 16 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.4 (xHigh) | 75 |
| 2 | Claude Opus 4.6 | 68.6 |
| 3 | Gemini 3.1 Pro (Preview) | 66.1 |
| 4 | GPT-5.4 (High) | 64.4 |
| 5 | Claude Sonnet 4.6 | 63.3 |
| 6 | GPT-5.4 (Medium) | 56.4 |
| 7 | GPT-5.4 (Low) | 55 |
| 8 | Gemini 3 Flash | 51.7 |
| 9 | Qwen 3.5 Omni Plus | 38.7 |
| 10 | MiniMax-M2.7 | 33.3 |
| 11 | MiMo-V2-Omni | 25.8 |
Interactive version: theaggregate.ai/benchmark?slug=sandboxed-coding-agents-are-competitive-omni-m · How It Works · Data refreshed daily, snapshot 2026-09-05.