Help keep public data accurate
See something missing, incorrect, or out of date? Tell us about this page.
See something missing, incorrect, or out of date?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 |
|---|---|---|---|---|---|---|
| DeepSeek V4.1 Flash 1017 | 17 Oct 2026 | 51 | ZeroBench-main with tools; official release changelog specifies Pass@5. Developer-reported percentage score (0–100), higher is better. DeepSeek-V4.1-Flash-1017 model card, release 2026-10-17, accessed 2026-10-17. Reasoning effort and any unlisted evaluation settings are not specified by this card. | Yes | Source | |
| DeepSeek V4.1 Flash | 10 Sept 2026 | 49 | Instruct model; reasoning_effort=100; Claude Code harness; 512k context; with tools; Pass@5 | Yes | Source | |
| Kimi K3 | 16 Jul 2026 | 23 | pass@5; official setting; run five times. | Yes | Source | |
| Seed 2.0 Pro | 14 Feb 2026 | 12 | Seed2 official benchmark table | ZeroBench (main) | Yes | Source | |
| Seed 2.0 Lite | 14 Feb 2026 | 8 | Seed2 official benchmark table | ZeroBench (main) | Yes | Source | |
| Seed 2.0 Mini | 14 Feb 2026 | 7 | Seed2 official benchmark table | ZeroBench (main) | Yes | Source |
