OpenAI
availableShows if the model has enough results for an index.o3
o3 is a reasoning model from OpenAI. 26 benchmarks count toward its score, in 7 categories.
IndexOverall score. 50 is the middle.53.6 ±3.7
CoverageShare of the index weight with results.95%
SpeedOutput tokens per second.77/s
Input / 1MUS dollars per 1M input tokens.$2 batch $1
Output / 1MUS dollars per 1M output tokens.$8 batch $4 US dollars per 1M output tokens in a batch.
ContextMaximum tokens in one request.200K
EloLMArena rating and rank.1410 (#130)
50 is the middle of the board. The range shows the doubt in the index. Batch work costs less.
58,579 votes. Elo shows what people prefer. It does not change the score.
CapabilitiesScore per category. 50 is the middle.
50 is the middleResults
26 counted| BenchmarkThe test name. | CategoryThe capability that the test measures. | ResultThe score from the publisher. | IndexThis result on the index scale. | RunThe settings of the run. | DateDate of the result. | Published byThe source of the result. |
|---|---|---|---|---|---|---|
| Artificial Analysis MATH-500 | Math | 99.2% | — | — | — | Artificial Analysis |
| MATH level 5 | Math | 97.8% | 46.7 | high effort | — | Epoch AI |
| MATH 500 | Math | 94.6% | 47.5 | — | 9 Jan 2026 | Vals AI |
| MGSM | Multilingual | 91.7% | — | — | 9 Jan 2026 | Vals AI |
| MMLU Pro | Knowledge | 85.6% | 55.4 | — | 1 Sept 2026 | Vals AI |
| AIME | Math | 85.3% | 53.5 | — | 16 Apr 2026 | Vals AI |
| GPQA Diamond | Knowledge | 84.1% | 55.9 | — | 1 Sept 2026 | Vals AI |
| LiveCodeBench | Coding | 83.9% | 60.6 | — | 1 Sept 2026 | Vals AI |
| OTIS Mock AIME 2024-2025 | Math | 83.9% | 58.5 | high effort | — | Epoch AI |
| Artificial Analysis GPQA Diamond | Knowledge | 82.7% | 53.3 | — | — | Artificial Analysis |
| GPQA diamond | Knowledge | 81.8% | 53.8 | high effort | — | Epoch AI |
| τ²-Bench Tool-Agent-User Evaluation | Agentic | 80.7% | 56.7 | — | — | Victor Barres et al. |
| MMMU Pro | Multimodal | 80.4% | 58.2 | — | 1 Sept 2026 | Vals AI |
| Artificial Analysis Long Context Reasoning | Reasoning | 74.7% | 59.9 | — | — | Artificial Analysis |
| τ²-bench Retail | Agentic | 73.9% | 51.8 | OpenAI | 30 Apr 2026 | Sierra Research |
| Artificial Analysis IFBench | Instruction | 71.4% | 61.5 | — | — | Artificial Analysis |
| Artificial Analysis MMMU-Pro | Multimodal | 70.1% | 52.8 | — | — | Artificial Analysis |
| SWE-Bench verified | Coding | 62.3% | 47.5 | medium effort | — | Epoch AI |
| ARC-AGI-1 (semi-private) | Reasoning | 60.8% | 55.9 | high effort | — | ARC Prize Foundation |
| SWE-bench Verified | Coding | 58.4% | 44.4 | mini-SWE-agent | 26 Feb 2026 | SWE-bench team |
| SWE-bench Verified | Coding | 58.4% | 44.4 | mini-SWE-agent | 1 Sept 2026 | SWE-bench team |
| τ²-bench Airline | Agentic | 52.0% | 36.1 | OpenAI | 2 Mar 2026 | Sierra Research |
| SimpleQA Verified | Knowledge | 49.4% | 67.3 | high effort | — | Epoch AI |
| Artificial Analysis Omniscience Accuracy | Knowledge | 38.6% | 61.6 | — | — | Artificial Analysis |
| SWE-bench Multimodal | Multimodal | 36.0% | — | GUIRepair | 17 Nov 2025 | SWE-bench team |
| SWE-bench Multimodal | Multimodal | 36.0% | — | GUIRepair | 17 Nov 2025 | SWE-bench team |
| Chess Puzzles | Reasoning | 34.0% | 67.5 | high effort | — | Epoch AI |
| FrontierMath-Tiers-1-3-v2-Private | Math | 33.3% | 50.2 | high effort | — | Epoch AI |
| Mystery Game Puzzles | Reasoning | 29.0% | 64.8 | high effort | — | Epoch AI |
| Artificial Analysis Intelligence Index | Knowledge | 20.2% | 47.7 | — | — | Artificial Analysis |
| Artificial Analysis Humanity's Last Exam | Knowledge | 20.1% | 46.4 | — | — | Artificial Analysis |
| FrontierMath-2025-02-28-Private | Math | 18.7% | 48.1 | high effort | — | Epoch AI |
| ARC-AGI-2 (semi-private) | Reasoning | 6.5% | 43.0 | high effort | — | ARC Prize Foundation |
| FrontierMath-Tier-4-2025-07-01-Private | Math | 2.1% | 45.7 | high effort | — | Epoch AI |
| Critical Physics Tasks | Reasoning | 1.1% | 40.7 | — | — | Artificial Analysis |
26 benchmarks count, from 31 of 35 results. A grey row does not count. Too few models took that benchmark.