o4-mini
| Released | 2025-04-16 |
|---|---|
| Price per 1M | $1.10 in · $4.40 out |
| Overall score | 48.9 · rank 125 |
| Context window | 200,000 tokens |
| Provider | OpenAI |
| Weights | Proprietary |
Scores
| Dimension | Score | Rank | Benchmarks |
|---|---|---|---|
| Overall: Balanced | 48.9 | 125 | 26 |
| Overall: Coding | 48.9 | 118 | 14 |
| Overall: Reasoning | 50.2 | 111 | 16 |
| Overall: Agentic work | 48.1 | 128 | 14 |
| Overall: Frontend | 49.8 | 89 | 7 |
| Coding | 49.3 | 120 | 3 |
| Agents | 47.1 | 127 | 2 |
| Reasoning | 49.8 | 114 | 9 |
| Math | 52.9 | 89 | 4 |
| Knowledge | 46.5 | 108 | 3 |
| General | 47.2 | 132 | 1 |
| Multimodal | 51.0 | 69 | 4 |
| Value vs price | −0.8 | 127 |
Configurations
| Configuration | Balanced | Benchmarks |
|---|---|---|
| Unspecified setting | 47.8 ±4.3 | 4 |
| High | 49.6 ±2.8 | 20 |
| Low | 46.7 ±4.3 | 7 |
| Medium | 49.5 ±2.9 | 12 |
Evidence
Coding 49.3 ±4.1
| Benchmark | Result | Source | Date |
|---|---|---|---|
| GSO | 3.6% | Epoch AI | 2026-11-11 |
| Aider Polyglot | 72.0% | Epoch AI | 2026-11-11 |
| LiveCodeBench (Vals run) | 82.2% | Vals AI | 2026-09-01 |
Agents 47.1 ±2.4
| Benchmark | Result | Source | Date |
|---|---|---|---|
| METR task suite | 63.9% | Epoch AI | 2026-11-11 |
| BFCL v4 | 53.2% | BFCL | 2026-04-12 |
Reasoning 49.8 ±2.1
| Benchmark | Result | Source | Date |
|---|---|---|---|
| ARC-AGI-2 | 6.1% | Epoch AI | 2026-11-11 |
| ARC-AGI-2 (ARC Prize verified) | 6.1% | ARC Prize | 2026-11-11 |
| CritPt | 0.6% | Epoch AI | 2026-11-11 |
| ARC-AGI-1 (ARC Prize verified) | 58.7% | ARC Prize | 2026-11-11 |
| ARC-AGI-1 | 58.7% | Epoch AI | 2026-11-11 |
| EnigmaEval | 9.2% | Epoch AI | 2026-11-11 |
| GPQA Diamond | 79.6% | Epoch AI | 2025-04-16 |
| GPQA Diamond (Vals run) | 74.5% | Vals AI | 2026-09-01 |
| Kagi Reasoning Benchmark | 67.6% | Kagi | 2026-11-11 |
| Chess Puzzles | 22.0% | Epoch AI | 2026-10-13 |
| SimpleBench | 38.7% | SimpleBench | 2026-11-11 |
| SimpleBench | 38.7% | Epoch AI | 2026-11-11 |
| Mystery Game Puzzles | 5.0% | Epoch AI | 2026-08-27 |
Math 52.9 ±1.2
| Benchmark | Result | Source | Date |
|---|---|---|---|
| FrontierMath Tiers 1-3 | 36.1% | Epoch AI | 2026-06-11 |
| MATH Level 5 | 97.8% | Epoch AI | 2025-04-16 |
| FrontierMath Tier 4 | 4.9% | Epoch AI | 2026-06-11 |
| OTIS Mock AIME 2024-2025 | 81.7% | Epoch AI | 2025-04-16 |
| FrontierMath Tiers 1-3 (2025 set) | 24.8% | Epoch AI | 2025-11-13 |
| FrontierMath Tier 4 (2025 set) | 6.3% | Epoch AI | 2025-07-01 |
Knowledge 46.5 ±3.4
| Benchmark | Result | Source | Date |
|---|---|---|---|
| SimpleQA Verified | 19.6% | Epoch AI | 2026-08-27 |
| Humanity's Last Exam | 18.1% | Epoch AI | 2026-11-11 |
| MMLU-Pro (Vals run) | 80.6% | Vals AI | 2026-09-01 |
General 47.2 ±4.1
| Benchmark | Result | Source | Date |
|---|---|---|---|
| LMArena Text | 1,391 | LMArena | 2026-11-09 |
Multimodal 51.0 ±2.8
| Benchmark | Result | Source | Date |
|---|---|---|---|
| LMArena Vision | 1,201 | LMArena | 2026-11-09 |
| MMMU-Pro (Vals run) | 79.7% | Vals AI | 2026-09-01 |
| VPCT | 57.5% | Epoch AI | 2026-11-11 |
| GeoBench | 64.0% | Epoch AI | 2026-11-11 |
Results not used in the score
| Benchmark | Result | Source | Why not used |
|---|---|---|---|
| ARC-AGI-1 | 41.8% | Epoch AI | A stronger configuration result was used |
| ARC-AGI-1 | 21.3% | Epoch AI | A stronger configuration result was used |
| ARC-AGI-1 (ARC Prize verified) | 41.8% | ARC Prize | A stronger configuration result was used |
| ARC-AGI-1 (ARC Prize verified) | 21.3% | ARC Prize | A stronger configuration result was used |
| ARC-AGI-2 | 2.4% | Epoch AI | A stronger configuration result was used |
| ARC-AGI-2 | 1.7% | Epoch AI | A stronger configuration result was used |
| ARC-AGI-2 (ARC Prize verified) | 2.4% | ARC Prize | A stronger configuration result was used |
| ARC-AGI-2 (ARC Prize verified) | 1.7% | ARC Prize | A stronger configuration result was used |
| BFCL v4 | 50.3% | BFCL | A stronger configuration result was used |
| Chess Puzzles | 15.0% | Epoch AI | A stronger configuration result was used |
| Chess Puzzles | 13.0% | Epoch AI | A stronger configuration result was used |
| EnigmaEval | 6.8% | Epoch AI | A stronger configuration result was used |
| FrontierMath Tier 4 (2025 set) | 2.1% | Epoch AI | A stronger configuration result was used |
| FrontierMath Tiers 1-3 | 28.8% | Epoch AI | A stronger configuration result was used |
| FrontierMath Tiers 1-3 | 16.1% | Epoch AI | A stronger configuration result was used |
| FrontierMath Tiers 1-3 (2025 set) | 19.0% | Epoch AI | A stronger configuration result was used |
| FrontierMath Tiers 1-3 (2025 set) | 10.7% | Epoch AI | A stronger configuration result was used |
| GPQA Diamond | 77.8% | Epoch AI | A stronger configuration result was used |
| GPQA Diamond | 75.3% | Epoch AI | A stronger configuration result was used |
| GeoBench | 64.0% | Epoch AI | A stronger configuration result was used |
| Humanity's Last Exam | 14.3% | Epoch AI | A stronger configuration result was used |
| OTIS Mock AIME 2024-2025 | 73.3% | Epoch AI | A stronger configuration result was used |
| OTIS Mock AIME 2024-2025 | 57.8% | Epoch AI | A stronger configuration result was used |
| SimpleQA Verified | 18.8% | Epoch AI | A stronger configuration result was used |
Names matched to this model
o4-mini, o4-mini-2025-04-16