See something missing, incorrect, or out of date? Tell us about this page.
AI benchmark results and model performance
Individual benchmark scores plotted by date.
| Organisation | Models | Reported | Top score | Info | Self reported | Source |
|---|---|---|---|---|---|---|
| Claude Opus 5.5 | 22 Sept 2026 | 18.20 | AgentArena (Net Improvement); Qwen4-Max launch chart (Nov 2026); score cited by Qwen from LM Arena; chart: https://pbs.twimg.com/media/HXNgxQIaEAAHHwZ.jpg | No | Source | |
| GPT-6 Astra | 03 Sept 2026 | 13.90 | AgentArena (Net Improvement); Qwen4-Max launch chart (Nov 2026); score cited by Qwen from LM Arena; chart: https://pbs.twimg.com/media/HXNgxQIaEAAHHwZ.jpg | No | Source | |
| GPT-6.1 Sol | 29 Sept 2026 | 11.80 | AgentArena (Net Improvement); Qwen4-Max launch chart (Nov 2026); score cited by Qwen from LM Arena; chart: https://pbs.twimg.com/media/HXNgxQIaEAAHHwZ.jpg | No | Source | |
| Qwen 4 Max | 16 Nov 2026 | 7.50 | AgentArena (Net Improvement); Qwen4-Max launch chart (Nov 2026); score cited by Qwen from LM Arena; chart: https://pbs.twimg.com/media/HXNgxQIaEAAHHwZ.jpg | No | Source | |
| Kimi K3.1 | - | 3.90 | AgentArena (Net Improvement); Qwen4-Max launch chart (Nov 2026); score cited by Qwen from LM Arena; chart: https://pbs.twimg.com/media/HXNgxQIaEAAHHwZ.jpg | No | Source |