o3-mini
o3-mini (OpenAI) was released on Jan 31, 2025. In the Epoch AI capability ranking (ECI), it ranks #121 out of 267 models, with 140.33. On Arena, which measures human preference in blind comparisons, it has 1,379 ELO points (#198). Against models of the same generation, its best result is on SciCode: 42.8%, #1 of 15.
Where o3-mini stands out and where it falls behind, by the numbers
o3-mini has published results on 25 tests compiled by Epoch AI. The yardstick here is its generation: its position among the models released between May 6, 2024 and May 1, 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 25 tests, one by one
| Test | Score | In its generation | Overall | Test record |
|---|---|---|---|---|
| GPQA Diamond | 77.0% | #4 / 52 | #90 / 173 | 96.4% Gemini 4 Argon |
| AIME 2024 and 2025 (OTIS) | 76.9% | #3 / 48 | #89 / 168 | 100.0% GPT-5.5 |
| Dtbench | 68.8% | #4 / 37 (tied with 1) | #91 / 148 | 98.9% Claude Opus 5.5 |
| Lmca | 18.97 | #6 / 26 | #93 / 121 | 68.23 Claude Opus 5.5 |
| CritPt | 0.3% | #3 / 18 (tied with 1) | #84 / 112 | 32.3% GPT-5.6 Sol |
| WeirdML | 43.7% | #4 / 26 | #63 / 112 | 93.6% GPT-6 Astra |
| SciCode | 42.8% | #1 / 15 | #70 / 104 | 66.9% Claude Opus 5.5 |
| Frontiermath Tiers 1 3 V2 | 18.6% | #3 / 8 | #76 / 86 | 93.7% GPT-6 Astra |
| ARC-AGI 2 | 3.0% | #3 / 16 | #64 / 83 | 95.0% GPT-6 Astra |
| ARC-AGI | 34.5% | #3 / 14 | #64 / 81 | 98.5% Claude Fable 5 |
| SimpleQA Verified | 15.3% | #6 / 9 | #73 / 81 | 76.0% Gemini 4 Argon |
| Chess puzzles | 15.0% | #3 / 9 | #64 / 80 | 76.0% GPT-6 Astra |
| Game puzzles | 7.0% | #3 / 5 (tied with 1) | #69 / 77 | 84.0% GPT-6 Astra |
| SimpleBench | 22.8% | #14 / 21 | #66 / 74 | 81.9% Claude Fable 5 |
| MATH level 5 | 96.5% | #3 / 43 | #8 / 73 | 98.1% GPT-5 |
| Frontiermath Tier 4 V2 | 0.0% | #2 / 2 | #65 / 67 | 100.0% GPT-6.1 Sol |
| FrontierMath | 12.4% | #3 / 21 | #35 / 65 | 51.7% GPT-5.5 |
| FrontierMath tier 4 | 4.2% | #2 / 6 | #23 / 48 | 39.6% GPT-5.5 Pro |
| Aider Polyglot | 60.40 | #4 / 23 (tied with 1) | #13 / 38 | 88.00 GPT-5 |
| Creative writing (Lech Mazur) | 6.17 | #21 / 22 | #37 / 38 | 8.60 GPT-5 |
| LiveBench | 75.88 | #2 / 26 | #4 / 32 | 82.35 Gemini 2.5 Pro (Jun 2025) |
| Fiction.liveBench | 43.8% | #10 / 14 | #22 / 31 | 100.0% o3 |
| Gso | 1.3% | #5 / 6 | #27 / 28 | 88.2% Claude Fable 5.1 |
| Cybench | 22.5% | #1 / 8 | #9 / 19 | 93.0% Claude Opus 4.6 |
| CAD-Eval | 54.0% | #4 / 12 (tied with 1) | #5 / 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 o3-mini 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 3.7 Sonnet | Anthropic | 141.14 | 1,360 | n/a | Feb 24, 2025 |
| GLM-4.6 | Z.ai (Zhipu AI) | 140.78 | 1,500 | n/a | Sep 30, 2025 |
| Grok-3 mini | xAI | 140.33 | 1,394 | n/a | Jun 24, 2025 |
| o3-mini | OpenAI | 140.33 | 1,379 | n/a | Jan 31, 2025 |
| Kimi K2 Instruct | Moonshot | 140.13 | n/a | n/a | Jul 12, 2025 |
| DeepSeek-V3.1 | DeepSeek | 139.91 | 1,469 | n/a | Aug 21, 2025 |
| Qwen3-30B-A3B-Thinking (Jul 2025) | Alibaba | 139.61 | 1,347 | n/a | Jul 30, 2025 |
o3-mini or Grok-3 mini?
Grok-3 mini is the model from another lab closest to o3-mini in capability. Across the 10 tests both took, this is the score:
| Test | o3-mini | Grok-3 mini |
|---|---|---|
| ARC-AGI 2 | 3.0% | 0.4% |
| FrontierMath | 12.4% | 5.9% |
| ARC-AGI | 34.5% | 16.5% |
| Fiction.liveBench | 43.8% | 65.6% |
| Aider Polyglot | 60.40 | 49.30 |
| Creative writing (Lech Mazur) | 6.17 | 7.35 |
| MATH level 5 | 96.5% | 90.9% |
| WeirdML | 43.7% | 42.6% |
| AIME 2024 and 2025 (OTIS) | 76.9% | 77.8% |
| GPQA Diamond | 77.0% | 76.3% |
Unweighted score: each test counts one point, whether math or coding. The dot marks who did better on each row.
What o3-mini costs to deliver each task
Price per token is what is on the price sheet; cost per task is what lands on the invoice, because it depends on how many tokens the model spends. Each point is an effort level, published by whoever ran the test and compiled by Epoch AI. The further up and to the left, the more score per dollar.
ARC-AGI 2
The score scale does not start at zero.
ARC-AGI
The score scale does not start at zero.
Other models from OpenAI
The models from the same lab closest in time, newest first.
| Model | Release | ECI | ELO | LiveBench | Access |
|---|---|---|---|---|---|
| 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 |
| o1 | Dec 17, 2024 | 141.84 | 1,383 | n/a | API access |
Have you used o3-mini? 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 o3-mini
Where does o3-mini rank among AI models?
In the ROO3 AI Benchmark, ordered by the Epoch AI ECI, o3-mini ranks #121 out of 267 models, with an ECI of 140.33. 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 o3-mini better than Grok-3 mini?
Across the 10 tests both models took and Epoch AI compiled, o3-mini 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 o3-mini 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 o3-mini are: SciCode (#1 of 15); Cybench (#1 of 8); LiveBench (#2 of 26). 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