Gemini 4 Argon
Claude Opus 5.5
Muse Wonder
MiMo-V2.6-Pro
Qwen3.8-Max
Kimi K3
GLM-5.3
ERNIE 5.1
DeepSeek-V4.1-Flash-20260910
Grok 4.20 Beta
amazon-nova-experimental-chat
Dola Seed 2.0 Pro
NVIDIA-Nemotron-3-Ultra-550B-A55B
Gemma 4 31B
Step 5 Preview
o3
o3 is a proprietary model from OpenAI, released Apr 16, 2025. It ranks #141 of 351 on the LMArena text leaderboard with a rating of 1410. It costs $2 per million input tokens and $8 per million output tokens, with a 200K-token context window.
- Developer
- OpenAI
- Released
- Apr 16, 2025
- Context window
- 200K
- Price in / out, per 1M tokens
- $2 / $8
- Weights
- Proprietary
- Inputs
- Images, Text, Files
- Reasoning
- Yes
- Tool use
- Yes
The rating LMArena published at each update since Apr 23, 2025. Ratings move as votes come in and new models join.
Benchmarks
| Benchmark | Score bar | Accuracy | Rank | Cost / test |
|---|---|---|---|---|
| LiveCodeBenchCoding | 83.91% | 45 / 143 | $2 / $8 | |
| MedCodeMedicine | 47.29% | 34 / 107 | $2 / $8 | |
| MedQAMedicine | 96.06% | 6 / 92 | $2 / $8 | |
| GPQA DiamondReasoning | 84.09% | 54 / 138 | $2 / $8 | |
| MMLU-ProReasoning | 85.59% | 56 / 138 | $2 / $8 | |
| AIMEReasoning | 85.28% | 37 / 96 | $2 / $8 | |
| MMMUReasoning | 80.42% | 44 / 93 | $2 / $8 |
Vals AI, as of Oct 24, 2026.
LMArena, by category
| Leaderboard | Rating | Rank | Votes |
|---|---|---|---|
| TextLMArena | 1410 | 141 / 351 | 58.9K |
| CodingLMArena | 1407 | 167 / 346 | 11.6K |
| MathLMArena | 1426 | 109 / 339 | 3.7K |
| Creative writingLMArena | 1358 | 157 / 349 | 7.6K |
| Instruction followingLMArena | 1368 | 171 / 351 | 15.5K |
| Hard promptsLMArena | 1402 | 163 / 351 | 25.5K |
| Multi-turnLMArena | 1405 | 146 / 349 | 9.7K |
| Long queriesLMArena | 1372 | 182 / 330 | 11.5K |
| ExpertLMArena | 1402 | 155 / 305 | 3K |
| Legal and governmentIndustries | 1428 | 114 / 325 | 3.7K |
| Medicine and healthcareIndustries | 1448 | 78 / 325 | 3.3K |
| Business and financeIndustries | 1393 | 153 / 344 | 9.5K |
| ChineseLanguages | 1437 | 144 / 328 | 3.9K |
| SpanishLanguages | 1395 | 153 / 249 | 1.2K |
| GermanLanguages | 1420 | 97 / 261 | 1.5K |
| FrenchLanguages | 1430 | 120 / 252 | 826 |
| JapaneseLanguages | 1403 | 75 / 236 | 1.4K |
| VisionLMArena | 1214 | 86 / 146 | 45.6K |
LMArena, CC BY 4.0, as of Oct 22, 2026. Where a model is rated at several effort levels, its best is shown.
Compare o3
OpenAI
The lab behind GPT and ChatGPT, which brought large language models to hundreds of millions of people. Founded 2015, based in San Francisco. openai.com
Questions
Who makes o3?
o3 is made by OpenAI, founded 2015 and based in San Francisco.
How much does o3 cost?
$2 per million input tokens and $8 per million output tokens, at its provider’s list price.
What is o3’s context window?
200K tokens (200,000).
How does o3 rank?
#141 of 351 on LMArena’s text leaderboard, rated 1410 (95% interval 1406–1414) from 58,894 votes, as of Oct 22, 2026.
Is o3 open source?
No. o3 is proprietary and served through its provider’s API.