o3
o3 is an API-only (closed weights) language model from OpenAI, released April 16, 2025. The context window is 195K tokens. On the Epoch Capabilities Index it scores 147.1, rank 83 of 256 rated models. API price: $2 input / $8 output per million tokens.
Check it against your own PC →
Lab
OpenAI
Released
April 16, 2025
Weights
API-only (closed weights)
Parameters
–
Context
195K
Price / M tokens
$2 / $8 in / out
Epoch ECI
147.1 #83
Strengths by category
ModelTape's 0–100 fit on Epoch AI benchmark results.
| Category | Score | Based on |
|---|---|---|
| Reasoning | 17.3 | Chess Puzzles, SimpleBench, Mystery Game Puzzles, Humanity's Last Exam |
| Math | 39.7 | MATH level 5, OTIS Mock AIME 2024-2025, FrontierMath-Tier-4-2025-07-01-Private, FrontierMath-Tiers-1-3-v2-Private |
| Coding | 45.4 | Aider polyglot, GSO-Bench, SWE-bench Verified |
| Knowledge | 71.3 | GPQA diamond, SimpleQA Verified |
Benchmark results
| Benchmark (Epoch AI) | Score |
|---|---|
| MATH level 5 | 97.8% |
| OTIS Mock AIME | 84.4% |
| Aider Polyglot | 81.3% |
| GPQA Diamond | 78.9% |
| SWE-bench Verified | 62.3% |
| SimpleQA Verified | 49.4% |
| SimpleBench | 43.7% |
| FrontierMath (Tiers 1–3) | 35.1% |
| Chess Puzzles | 33.7% |
| OSWorld | 23.0% |
| Mystery Game Puzzles | 21.8% |
| Humanity's Last Exam | 20.3% |
| GSO-Bench | 8.8% |
| FrontierMath-Tier-4-2025-07-01-Private | 3.5% |
Links
Models with a similar score
GPT-5.1 · Beam 1.0 · Inkling-Small · GLM-5.1 · DeepSeek-V4-Pro · GPT-5 · Grok 4.3 Beta · Qwen 3.6 Max (Preview)
More from OpenAI
GPT-6.1 Terra · GPT-6.1 Sol · GPT-6 Sol · GPT-6 Luna · GPT-6 Astra · GPT-5.6 Cyber · Daybreak Red · Daybreak Blue