Automated page. Written and translated automatically by AI and checked against the sources by software; no human editor. SI-IQ is an index built from public benchmark results, not an IQ test. How it works
How smart is o3?
o3 is a model from OpenAI, released on Apr 16, 2025. Its SI-IQ is 116 (90% interval 113–119), estimated from 10 benchmark results by 3 independent evaluators. Its strongest area is Reasoning and its weakest is Coding.
Facts
- Developer
- OpenAI
- Released
- Apr 16, 2025
- Access
- API access
- Benchmarks used
- 10 results from 3 evaluators
- API price
- $2.00 input / $8.00 output per 1M tokens (developer's price list, Oct 18, 2026)
- Context window
- 200,000 tokens
Ability by area (SI-IQ, 90% interval)
Every benchmark result we used
Each score is one value exactly as the source published it: for Epoch AI the best of its run settings, for OpenRouter the median hosting provider (the developer's own hosting when it ran the test). “Implied SI-IQ” is the ability that this single result points to; the overall SI-IQ is their precision-weighted average.
| Benchmark | Score | Evaluated by | Source |
|---|---|---|---|
| Chess Puzzles | 37.0% | Epoch AI | Epoch AI Benchmarking Hub · data file |
| Mystery Game Puzzles | 29.0% | Epoch AI | Epoch AI Benchmarking Hub · data file |
| FrontierMath Tiers 1–3 | 33.3% | Epoch AI | Epoch AI Benchmarking Hub · data file |
| MATH Level 5 | 97.8% | Epoch AI | Epoch AI Benchmarking Hub · data file |
| OTIS Mock AIME 2024–2025 | 84.4% | Epoch AI | Epoch AI Benchmarking Hub · data file |
| Aider Polyglot | 81.3% | Aider | Aider LLM Leaderboards · data file |
| SWE-bench Verified (Epoch AI run) | 62.3% | Epoch AI | Epoch AI Benchmarking Hub · data file |
| GPQA Diamond | 81.8% | Epoch AI | Epoch AI Benchmarking Hub · data file |
| SimpleQA Verified | 49.4% | Epoch AI | Epoch AI Benchmarking Hub · data file |
| Berkeley Function Calling Leaderboard | 63.0% | UC Berkeley (Gorilla) | Berkeley Function Calling Leaderboard · data file |
Human preference (Arena)
On the Arena leaderboard, where people vote between two anonymous answers, o3 has a rating of 1,432 (1,428–1,435) from 58,908 votes (published Oct 15, 2026). Human preference is a different measure from benchmark ability and is not part of SI-IQ. Arena (Hugging Face)
Models with similar SI-IQ
| Model | SI-IQ | |
|---|---|---|
| GLM-5.1Z.ai (Zhipu AI) | 116 | o3 vs GLM-5.1 |
| DeepSeek-V4-ProDeepSeek | 117 | o3 vs DeepSeek-V4-Pro |
| GPT-5.4 MiniOpenAI | 117 | o3 vs GPT-5.4 Mini |
| GPT-5 miniOpenAI | 115 | o3 vs GPT-5 mini |
| Qwen3.6 27BAlibaba | 115 | o3 vs Qwen3.6 27B |
Computed with SI-IQ method version 1.0.0. Read the full methodology
Data sources: Aider polyglot leaderboard data (Apache-2.0); Arena leaderboard dataset on Hugging Face (CC BY 4.0); Berkeley Function Calling Leaderboard, Gorilla project (Apache-2.0); Epoch AI, ‘Capabilities & benchmarking’ (CC BY 4.0); API list prices via LiteLLM's price table (MIT); each price links to the developer's pricing page.