Grok 4
Grok 4 (xAI) was released on Jul 9, 2025. In the Epoch AI capability ranking (ECI), it ranks #78 out of 267 models, with 146.43. Against models of the same generation, its best result is on GPQA Diamond: 87.0%, #1 of 55.
Where Grok 4 stands out and where it falls behind, by the numbers
Grok 4 has published results on 19 tests compiled by Epoch AI. The yardstick here is its generation: its position among the models released between Oct 12, 2024 and Oct 7, 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 19 tests, one by one
| Test | Score | In its generation | Overall | Test record |
|---|---|---|---|---|
| GPQA Diamond | 87.0% | #1 / 55 | #59 / 173 | 96.4% Gemini 4 Argon |
| AIME 2024 and 2025 (OTIS) | 84.0% | #7 / 54 | #78 / 168 | 100.0% GPT-5.5 |
| WeirdML | 45.7% | #13 / 39 | #57 / 112 | 93.6% GPT-6 Astra |
| ARC-AGI 2 | 16.0% | #2 / 28 | #46 / 83 | 95.0% GPT-6 Astra |
| ARC-AGI | 66.7% | #2 / 28 | #45 / 81 | 98.5% Claude Fable 5 |
| SimpleBench | 60.5% | #3 / 28 | #25 / 74 | 81.9% Claude Fable 5 |
| FrontierMath | 19.7% | #4 / 31 | #30 / 65 | 51.7% GPT-5.5 |
| ForecastBench | 60.90 | #11 / 24 | #18 / 62 | 62.50 o3 |
| FrontierMath tier 4 | 2.1% | #11 / 18 (tied with 2) | #34 / 48 | 39.6% GPT-5.5 Pro |
| Aider Polyglot | 79.60 | #4 / 33 | #4 / 38 | 88.00 GPT-5 |
| Creative writing (Lech Mazur) | 7.69 | #18 / 34 | #20 / 38 | 8.60 GPT-5 |
| Terminal-Bench | 27.2% | #6 / 10 | #27 / 33 | 84.7% GPT-5.5 |
| BALROG | 43.6% | #1 / 11 | #9 / 32 | 68.3% GPT-6 Astra |
| Fiction.liveBench | 96.9% | #2 / 30 (tied with 1) | #2 / 31 | 100.0% o3 |
| METR: time horizon | 110 | #5 / 15 | #12 / 29 | 719 Claude Opus 4.6 |
| DeepResearch Bench | 47.3% | #4 / 10 | #11 / 24 | 55.3% Claude Opus 4.6 |
| GeoBench | 2,719 | #14 / 16 | #22 / 24 | 4,333 Gemini 3 Flash |
| Cybench | 43.0% | #2 / 11 | #4 / 19 | 93.0% Claude Opus 4.6 |
| GDPval | 21.1% | #7 / 7 | #10 / 11 | 49.7% GPT-5.2 |
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 Grok 4 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 Sonnet 4.5 | Anthropic | 146.83 | 1,488 | n/a | Sep 29, 2025 |
| MiniMax-M2.5 | MiniMax | 146.67 | 1,393 | n/a | Feb 12, 2026 |
| Grok 4 | xAI | 146.43 | n/a | n/a | Jul 9, 2025 |
| Qwen 3.5 Plus (hosted 397B-A17B) | Alibaba | 146.32 | n/a | n/a | Feb 16, 2026 |
| Qwen3.5-122B-A10B | Alibaba | 146.26 | 1,466 | n/a | Feb 24, 2026 |
| GPT-5.4 Nano | OpenAI | 146.12 | 1,404 | 69.6 | Mar 17, 2026 |
| DeepSeek-V3.2 | DeepSeek | 146.11 | 1,455 | n/a | Dec 1, 2025 |
Grok 4 or GPT-5.4 Nano?
GPT-5.4 Nano is the model from another lab closest to Grok 4 in capability. Across the 7 tests both took, this is the score:
| Test | Grok 4 | GPT-5.4 Nano |
|---|---|---|
| FrontierMath tier 4 | 2.1% | 6.3% |
| ARC-AGI 2 | 16.0% | 5.7% |
| FrontierMath | 19.7% | 25.9% |
| ARC-AGI | 66.7% | 51.5% |
| GPQA Diamond | 87.0% | 78.5% |
| WeirdML | 45.7% | 49.2% |
| AIME 2024 and 2025 (OTIS) | 84.0% | 87.8% |
Unweighted score: each test counts one point, whether math or coding. The dot marks who did better on each row.
Other models from xAI
The models from the same lab closest in time, newest first.
| Model | Release | ECI | ELO | LiveBench | Access |
|---|---|---|---|---|---|
| Grok 4.5 | Jul 8, 2026 | 153.93 | 1,495 | 75.8 | API access |
| Grok 4.3 (no thinking) | Apr 17, 2026 | 149.21 | 1,425 | 62.3 | Hosted access (no API) |
| Grok 4.3 Beta | Apr 17, 2026 | 148.95 | n/a | n/a | Hosted access (no API) |
| Grok 4.20 | Feb 17, 2026 | 152.08 | n/a | n/a | API access |
| Grok 4 Fast | Sep 19, 2025 | 144.19 | 1,437 | n/a | API access |
| Grok 4 | Jul 9, 2025 | 146.43 | n/a | n/a | API access |
| Grok-3 mini | Jun 24, 2025 | 140.33 | 1,394 | n/a | API access |
| Grok 3 | Apr 9, 2025 | 138.32 | n/a | n/a | API access |
| Grok-2 (Dec 2024) | Dec 12, 2024 | 130.49 | n/a | n/a | API access |
Have you used Grok 4? 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 Grok 4
Where does Grok 4 rank among AI models?
In the ROO3 AI Benchmark, ordered by the Epoch AI ECI, Grok 4 ranks #78 out of 267 models, with an ECI of 146.43. 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 Grok 4 better than GPT-5.4 Nano?
Across the 7 tests both models took and Epoch AI compiled, Grok 4 comes out ahead on 3, behind on 4 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 Grok 4 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 Grok 4 are: GPQA Diamond (#1 of 55); BALROG (#1 of 11); ARC-AGI 2 (#2 of 28). 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