GPT-4.1
GPT-4.1 (OpenAI) was released on Apr 14, 2025. In the Epoch AI capability ranking (ECI), it ranks #137 out of 267 models, with 136.77. Against models of the same generation, its best result is on ForecastBench: 61.50, #4 of 19.
Where GPT-4.1 stands out and where it falls behind, by the numbers
GPT-4.1 has published results on 23 tests compiled by Epoch AI. The yardstick here is its generation: its position among the models released between Jul 18, 2024 and Jul 13, 2025 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 23 tests, one by one
| Test | Score | In its generation | Overall | Test record |
|---|---|---|---|---|
| GPQA Diamond | 66.9% | #16 / 56 | #107 / 173 | 96.4% Gemini 4 Argon |
| AIME 2024 and 2025 (OTIS) | 38.3% | #22 / 53 | #118 / 168 | 100.0% GPT-5.5 |
| Dtbench | 68.3% | #10 / 41 | #93 / 148 | 98.9% Claude Opus 5.5 |
| Lmca | 25.61 | #7 / 29 | #84 / 121 | 68.23 Claude Opus 5.5 |
| WeirdML | 39.0% | #14 / 32 | #75 / 112 | 93.6% GPT-6 Astra |
| Frontiermath Tiers 1 3 V2 | 6.0% | #7 / 8 | #82 / 86 | 93.7% GPT-6 Astra |
| ARC-AGI 2 | 0.4% | #15 / 23 (tied with 1) | #74 / 83 | 95.0% GPT-6 Astra |
| ALE-Bench | 558 | #6 / 8 | #75 / 82 | 2,951 GPT-6 Astra |
| ARC-AGI | 5.5% | #15 / 21 (tied with 1) | #74 / 81 | 98.5% Claude Fable 5 |
| SimpleQA Verified | 31.1% | #3 / 8 | #60 / 81 | 76.0% Gemini 4 Argon |
| Chess puzzles | 3.0% | #5 / 8 (tied with 1) | #76 / 80 | 76.0% GPT-6 Astra |
| SimpleBench | 27.0% | #17 / 25 | #61 / 74 | 81.9% Claude Fable 5 |
| MATH level 5 | 83.0% | #16 / 44 | #22 / 73 | 98.1% GPT-5 |
| FrontierMath | 5.5% | #8 / 26 | #44 / 65 | 51.7% GPT-5.5 |
| ForecastBench | 61.50 | #4 / 19 | #7 / 62 | 62.50 o3 |
| FrontierMath tier 4 | 0.0% | #7 / 10 (tied with 3) | #40 / 48 | 39.6% GPT-5.5 Pro |
| Hle | 5.4% | #11 / 15 | #34 / 39 | 54.8% GPT-6 Astra |
| Aider Polyglot | 52.40 | #14 / 29 | #21 / 38 | 88.00 GPT-5 |
| EnigmaEval | 2.2% | #9 / 14 | #29 / 36 | 39.3% Claude Fable 5 |
| SWE-bench Verified | 48.5% | #5 / 5 | #31 / 32 | 83.5% Claude Opus 4.7 |
| Fiction.liveBench | 62.5% | #7 / 22 (tied with 2) | #12 / 31 | 100.0% o3 |
| GeoBench | 3,791 | #7 / 18 | #12 / 24 | 4,333 Gemini 3 Flash |
| CAD-Eval | 42.0% | #8 / 12 | #8 / 14 | 74.0% o3 |
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 GPT-4.1 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 |
|---|---|---|---|---|---|
| DeepSeek-R1-Distill-Qwen-32B | DeepSeek | 137.42 | n/a | n/a | Jan 20, 2025 |
| Qwen3-30B-A3B-Instruct (Jul 2025) | Alibaba | 137.42 | n/a | n/a | Jul 29, 2025 |
| GPT-4.1 | OpenAI | 136.77 | n/a | n/a | Apr 14, 2025 |
| Qwen3-8B | Alibaba | 136.17 | n/a | n/a | Apr 28, 2025 |
| DeepSeek-V3 (Mar 2025) | DeepSeek | 135.93 | 1,346 | n/a | Mar 24, 2025 |
| Mistral Medium 3.1 | Mistral AI | 135.67 | n/a | n/a | Aug 12, 2025 |
| Gemini 2.0 Flash Thinking Exp | Google DeepMind | 135.39 | n/a | n/a | Jan 21, 2025 |
GPT-4.1 or DeepSeek-V3 (Mar 2025)?
DeepSeek-V3 (Mar 2025) is the model from another lab closest to GPT-4.1 in capability. Across the 10 tests both took, this is the score:
| Test | GPT-4.1 | DeepSeek-V3 (Mar 2025) |
|---|---|---|
| FrontierMath | 5.5% | 1.7% |
| Lmca | 25.61 | 15.47 |
| MATH level 5 | 83.0% | 75.5% |
| WeirdML | 39.0% | 36.1% |
| Dtbench | 68.3% | 64.8% |
| Aider Polyglot | 52.40 | 55.10 |
| ForecastBench | 61.50 | 59.10 |
| AIME 2024 and 2025 (OTIS) | 38.3% | 37.8% |
| GPQA Diamond | 66.9% | 67.6% |
| SimpleBench | 27.0% | 27.2% |
Unweighted score: each test counts one point, whether math or coding. The dot marks who did better on each row.
Other models from OpenAI
The models from the same lab closest in time, newest first.
| Model | Release | ECI | ELO | LiveBench | Access |
|---|---|---|---|---|---|
| gpt-oss-120b | Aug 5, 2025 | 140.00 | 1,385 | n/a | Open weights (unrestricted) |
| o3-pro | Jun 10, 2025 | 147.37 | n/a | n/a | API access |
| o3 | Apr 16, 2025 | 146.87 | n/a | n/a | API access |
| o4-mini | Apr 16, 2025 | 145.60 | n/a | n/a | API access |
| GPT-4.1 | Apr 14, 2025 | 136.77 | n/a | n/a | API access |
| GPT-4.1 mini | Apr 14, 2025 | 135.00 | n/a | n/a | API access |
| GPT-4.1 nano | Apr 14, 2025 | 129.63 | n/a | n/a | API access |
| GPT-4.5 | Feb 27, 2025 | 136.73 | n/a | n/a | API access |
| o3-mini | Jan 31, 2025 | 140.33 | 1,379 | n/a | API access |
Have you used GPT-4.1? 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 GPT-4.1
Where does GPT-4.1 rank among AI models?
In the ROO3 AI Benchmark, ordered by the Epoch AI ECI, GPT-4.1 ranks #137 out of 267 models, with an ECI of 136.77. 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 GPT-4.1 better than DeepSeek-V3 (Mar 2025)?
Across the 10 tests both models took and Epoch AI compiled, GPT-4.1 comes out ahead on 7, behind on 3 and tied on 0. 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 GPT-4.1 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 GPT-4.1 are: ForecastBench (#4 of 19); Lmca (#7 of 29); Dtbench (#10 of 41). 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