DeepSeek V4.1 Flash
DeepSeek V4.1 Flash (DeepSeek) was released on Sep 9, 2026. In the Epoch AI capability ranking (ECI), it ranks #31 out of 267 models, with 155.57. On LiveBench, the overall average is 81.1, #7 out of 58 scored models. Against models of the same generation, its best result is on Game puzzles: 43.0%, #14 of 59, tied with 1 others.
Where DeepSeek V4.1 Flash stands out and where it falls behind, by the numbers
DeepSeek V4.1 Flash has published results on 22 tests compiled by Epoch AI. The yardstick here is its generation: its position among the models released between Dec 13, 2025 and Dec 8, 2026 that took the same test. Against the whole archive, which includes 2023 models, a recent release lands near the top of almost everything, and that says nothing.
Where it stands out
Tests where it sits in the top third of its own generation.
Where it falls behind
Tests where it sits in the bottom half of its own generation.
See all 22 tests, one by one
| Test | Score | In its generation | Overall | Test record |
|---|---|---|---|---|
| GPQA Diamond | 89.8% | #39 / 75 | #41 / 173 | 96.4% Gemini 4 Argon |
| AIME 2024 and 2025 (OTIS) | 98.3% | #22 / 77 (tied with 3) | #22 / 168 | 100.0% GPT-5.5 |
| Dtbench | 89.9% | #39 / 65 (tied with 1) | #42 / 148 | 98.9% Claude Opus 5.5 |
| Lmca | 46.97 | #28 / 65 | #28 / 121 | 68.23 Claude Opus 5.5 |
| CritPt | 14.3% | #39 / 77 (tied with 1) | #39 / 112 | 32.3% GPT-5.6 Sol |
| SciCode | 51.9% | #41 / 75 | #41 / 104 | 66.9% Claude Opus 5.5 |
| Frontiermath Tiers 1 3 V2 | 67.4% | #29 / 65 | #31 / 86 | 93.7% GPT-6 Astra |
| ARC-AGI 2 | 72.9% | #18 / 44 | #18 / 83 | 95.0% GPT-6 Astra |
| ALE-Bench | 1,092 | #32 / 59 | #36 / 82 | 2,951 GPT-6 Astra |
| WebDev Arena | 1,614 | #18 / 69 | #18 / 82 | 1,820 Claude Opus 5.5 |
| ARC-AGI | 94.5% | #15 / 44 (tied with 2) | #15 / 81 | 98.5% Claude Fable 5 |
| SimpleQA Verified | 42.9% | #37 / 61 | #42 / 81 | 76.0% Gemini 4 Argon |
| Chess puzzles | 48.0% | #16 / 60 (tied with 2) | #17 / 80 | 76.0% GPT-6 Astra |
| Game puzzles | 43.0% | #14 / 59 (tied with 1) | #14 / 77 | 84.0% GPT-6 Astra |
| ProofBench | 54.0% | #26 / 66 (tied with 1) | #26 / 74 | 100.0% Claude Fable 5.1 |
| Frontiermath Tier 4 V2 | 26.8% | #32 / 55 (tied with 5) | #34 / 67 | 100.0% GPT-6.1 Sol |
| Apex Agents | 39.5% | #38 / 48 | #38 / 50 | 82.4% Gemini 4 Argon |
| FrontierCode | 43.9% | #17 / 43 | #17 / 43 | 54.6% Claude Opus 5.5 |
| Gdp Pdf | 19.8% | #23 / 37 | #23 / 37 | 35.0% GPT-6.1 Sol |
| Furniture Assembly | 34.2% | #20 / 32 (tied with 1) | #21 / 34 | 83.3% Claude Opus 5.5 |
| Surface Evolver Bench | 46.3% | #17 / 23 | #17 / 24 | 95.0% Claude Fable 5 |
| Weirdml V3 | 5.9% | #10 / 10 | #11 / 11 | 42.2% GPT-6 Astra |
For each test, the model's best effort level counts, as in the ranking. Source of the scores: Epoch AI, under CC BY 4.0.
What to compare DeepSeek V4.1 Flash with
The models from other labs closest in capability and released in the same period, at most two per lab. Click any of them to see its page.
| Model | Lab | ECI | ELO | LiveBench | Release |
|---|---|---|---|---|---|
| Claude Opus 4.7 | Anthropic | 156.32 | 1,532 | 76.5 | Apr 16, 2026 |
| Grok 4.6 | xAI | 156.31 | 1,475 | 78.0 | Aug 12, 2026 |
| GPT-6 Luna | OpenAI | 156.31 | n/a | 72.0 | Sep 22, 2026 |
| GLM-5.3 | Z.ai (Zhipu AI) | 155.91 | n/a | 76.1 | Aug 14, 2026 |
| DeepSeek V4.1 Flash | DeepSeek | 155.57 | n/a | 81.1 | Sep 9, 2026 |
| Claude Opus 4.6 | Anthropic | 155.27 | 1,550 | 74.5 | Feb 5, 2026 |
| Muse Spark 1.2 | Meta AI | 154.90 | 1,509 | 78.0 | Aug 5, 2026 |
DeepSeek V4.1 Flash or Claude Opus 4.6?
Claude Opus 4.6 is the model from another lab closest to DeepSeek V4.1 Flash in capability. Across the 17 tests both took, this is the score:
| Test | DeepSeek V4.1 Flash | Claude Opus 4.6 |
|---|---|---|
| Chess puzzles | 48.0% | 21.0% |
| Game puzzles | 43.0% | 25.0% |
| FrontierCode | 43.9% | 26.6% |
| Furniture Assembly | 34.2% | 28.3% |
| Lmca | 46.97 | 55.76 |
| Apex Agents | 39.5% | 46.3% |
| ALE-Bench | 1,092 | 997 |
| SimpleQA Verified | 42.9% | 47.0% |
| ProofBench | 54.0% | 50.0% |
| AIME 2024 and 2025 (OTIS) | 98.3% | 91.1% |
Unweighted score: each test counts one point, whether math or coding. The dot marks who did better on each row.
Other models from DeepSeek
The models from the same lab closest in time, newest first.
| Model | Release | ECI | ELO | LiveBench | Access |
|---|---|---|---|---|---|
| DeepSeek V4.1 Flash | Sep 9, 2026 | 155.57 | n/a | 81.1 | Open weights (unrestricted) |
| DeepSeek V4 Pro 0813 | Aug 13, 2026 | 155.29 | n/a | 77.4 | Open weights (unrestricted) |
| DeepSeek V4 Flash 0731 | Jul 31, 2026 | 154.49 | n/a | 74.2 | Open weights (unrestricted) |
| DeepSeek-V4-Pro | Apr 24, 2026 | 149.19 | 1,486 | 71.6 | Open weights (unrestricted) |
| DeepSeek v4 (unknown) | Apr 24, 2026 | 146.26 | n/a | n/a | Open weights (unrestricted) |
| DeepSeek-V4-Flash | Apr 24, 2026 | 146.05 | 1,468 | 65.5 | Open weights (unrestricted) |
| DeepSeek-V3.2 | Dec 1, 2025 | 146.11 | 1,455 | n/a | Open weights (unrestricted) |
| DeepSeek-V3.2-Exp | Sep 29, 2025 | 144.97 | 1,461 | n/a | Open weights (unrestricted) |
| DeepSeek-V3.1 | Aug 21, 2025 | 139.91 | 1,469 | n/a | Open weights (unrestricted) |
Have you used DeepSeek V4.1 Flash? Rate it
Users' opinion, not measurement: the rating is kept apart from the ranking and does not move anyone's position. One vote per browser, no sign-up, and you can change your rating later.
What people ask about DeepSeek V4.1 Flash
Where does DeepSeek V4.1 Flash rank among AI models?
In the ROO3 AI Benchmark, ordered by the Epoch AI ECI, DeepSeek V4.1 Flash ranks #31 out of 267 models, with an ECI of 155.57. The ECI is a composite index that combines several hard benchmarks into a single scale. The position changes when a new model comes out, and the page is refreshed six times a day.
Is DeepSeek V4.1 Flash better than Claude Opus 4.6?
Across the 17 tests both models took and Epoch AI compiled, DeepSeek V4.1 Flash comes out ahead on 12, behind on 4 and tied on 1. It is a test-by-test count with no weighting: a win in math counts the same as a win in coding. To choose between the two, look at the tests that resemble your own use.
What is DeepSeek V4.1 Flash best at?
Against models of the same generation (released up to 9 months before and up to 3 months after it), the best relative results of DeepSeek V4.1 Flash are: Game puzzles (#14 of 59); WebDev Arena (#18 of 69); Chess puzzles (#16 of 60). The yardstick is its position among the models of that period that took the same test.
Where do the numbers on this page come from?
From the same three sources as the ranking, all openly licensed: Epoch AI (ECI and individual tests, CC BY 4.0), Arena (human-preference ELO, CC BY 4.0) and LiveBench (overall average, CC BY-SA 4.0). ROO3 does not measure any model. When the page shows a figure published by the maker itself, it appears in the written analysis, labeled as the maker's, and never enters the ranking.
Where each number comes from
Epoch AI, "AI Benchmarking Hub". Published at epoch.ai. Data under the CC BY 4.0 licence, reordered by roo3.co.
The Arena (LMArena) leaderboard, from the public lmarena-ai/leaderboard-dataset on Hugging Face, under the CC BY 4.0 licence. We are not affiliated with the Arena.
LiveBench, by the LiveBench team, published at livebench.ai under the CC BY-SA 4.0 license. The overall average is recomputed by roo3.co with the same formula as the website (average of the categories), and those numbers stay under CC BY-SA 4.0. We are not affiliated with LiveBench.
← See the full ranking