Agnes 3.0 Flash vs DeepSeek V4.1 Flash 1017
On the 10 benchmarks both have, DeepSeek V4.1 Flash 1017 scores higher on 9 and Agnes 3.0 Flash on 1.
No benchmark publisher we read lists Agnes 3.0 Flash any more, so its figures stopped updating on 22 Sep 2026. DeepSeek V4.1 Flash 1017's figures are current as of 13 Nov 2026. Each score shows the date it was published.
| Benchmark |  Agnes 3.0 Flash |  DeepSeek V4.1 Flash 1017 (max) |
|---|---|---|
| AA Intelligence Index v4.3.2 | 35.5 11 Sep 2026 | 45.1 17 Oct 2026 |
| AA-LCR v1.1 | 81.0% 11 Sep 2026 | 82.7% 17 Oct 2026 |
| AA-Omniscience v1.1 | -10.6 11 Sep 2026 | 0.9 17 Oct 2026 |
| Humanity's Last Exam (AA) | 38.5% 11 Sep 2026 | 43.3% 17 Oct 2026 |
| SciCode (AA) | 51.6% 11 Sep 2026 | 55.9% 17 Oct 2026 |
| CritPt | 15.1% 19 Sep 2026 | 24.0% 17 Oct 2026 |
| GDPval-AA v2.1, normalised score | 52.5% 20 Sep 2026 | 62.3% 17 Oct 2026 |
| Terminal-Bench 4.0 (AA) | 7.1% 22 Sep 2026 | 33.8% 17 Oct 2026 |
| AA-Omniscience v1.1: accuracy | 25.5% 22 Sep 2026 | 39.6% 17 Oct 2026 |
| AA-Omniscience v1.1: non-hallucination | 51.6% 22 Sep 2026 | 35.9% 17 Oct 2026 |
| Blended price (per 1M tokens, 3:1) | $0.075 | $0.525 |
| Output speed | 317 tok/s | 158 tok/s |
See everything we last measured for Agnes 3.0 Flash, or DeepSeek V4.1 Flash 1017's page for how it compares with models that are still measured. The full list is on the leaderboards.