OpenAI o3
OpenAI o3 is a reasoning model for complex analytical, coding, mathematical, and scientific tasks.
- Reasoning
Added Apr 16, 2025
Pricing
Auto routing · per 1M tokens- Input
- $2.00
- Output
- $8.00
- Cache read
- $1.00
Specifications
- Context window
- 200K
- Max output
- 100K
Benchmarks
Benchmarks
Sourced from LMArena.
Arena Score
1431.5
Overall Rank
#114 / 421
Votes
58,811
Confidence Interval
1427.9 - 1435.1
Category Scores
Coding
#144 / 416
11,559 votes
1460.2
Math
#84 / 404
3,686 votes
1447.9
Longer Query
#161 / 399
11,511 votes
1411.2
Creative Writing
#143 / 419
7,620 votes
1383.1
Instruction Following
#147 / 421
15,479 votes
1402.6
Hard Prompts
#134 / 421
25,479 votes
1441.2
Additional Categories22
Polish
#67 / 244
3,833 votes
1457.7
Industry Medicine And Healthcare
#68 / 390
3,326 votes
1475.7
Japanese
#68 / 288
1,417 votes
1424.5
Korean
#91 / 289
1,198 votes
1393.7
German
#93 / 320
1,510 votes
1439.9
Industry Legal And Government
#94 / 393
3,738 votes
1452.1
French
#96 / 301
824 votes
1459.9
Industry Mathematical
#96 / 401
3,437 votes
1448.4
Industry Life And Physical And Social Science
#102 / 419
9,915 votes
1455.8
Non English
#107 / 421
31,528 votes
1422.3
Russian
#110 / 385
4,674 votes
1429.3
Exclude Ties
#115 / 421
42,811 votes
1424.8
English
#121 / 421
27,231 votes
1439.2
Industry Business And Management And Financial Operations
#121 / 414
9,470 votes
1426.6
Spanish
#122 / 303
1,219 votes
1417.3
Chinese
#124 / 395
3,878 votes
1462.4
Industry Entertainment And Sports And Media
#124 / 419
10,486 votes
1394.2
Expert
#125 / 371
3,008 votes
1448.2
Hard Prompts English
#129 / 419
12,081 votes
1449.9
Industry Software And It Services
#137 / 421
19,110 votes
1455.0
Industry Writing And Literature And Language
#137 / 420
13,226 votes
1396.1
Multi Turn
#137 / 419
9,643 votes
1420.2
Published 2026-11-09 · Matched as o3 (2025-04-16) · CC BY 4.0
LMArena DatasetLeaderboard results
| Benchmark | Score | Rank on NanoGPT |
|---|---|---|
Epoch AI Benchmarking Hub Epoch AI data, CC BY 4.0 · Updated Nov 13, 2026 | ||
Epoch Capabilities Index | 147.1 | #74 of 115 |
FrontierMath v2 (Tiers 1–3) | 33.3% | #67 of 83 |
SWE-bench Verified | 62.3% | #24 of 28 |
GPQA Diamond | 81.8% | #79 of 119 |
SimpleQA Verified | 49.4% | #31 of 82 |
OTIS Mock AIME 2024–2025 | 84.4% | #75 of 121 |
Chess Puzzles | 37.0% | #42 of 82 |
Mystery Game Puzzles | 29.0% | #36 of 77 |
Creative Writing v3 | 1675 | #37 of 96 |
Longform Writing | 62.7 | #52 of 97 |
EQ-Bench 3 | 1500 | #20 of 49 |
UGI Leaderboard UGI Leaderboard data, Apache-2.0 · Updated Oct 16, 2026 | ||
UGI Score | 52.1 | #32 of 149 |
UGI Willingness | 2.2 | #114 of 149 |
UGI Writing | 66.8 | #21 of 149 |
Filtered and ranked to models on NanoGPT.
Providers
Auto routing is available for this model. Explicit provider selection is not available.
 Recommended | $2.00 | $8.00 | $1.00 | N/A | N/A |
