GPT 5 Nano
Optimized for speed and ideal for applications requiring low latency. The fastest and most efficient GPT-5 variant.
- Reasoning
- Vision
Added Aug 7, 2025
Pricing
Auto routing · per 1M tokens- Input
- $0.05
- Output
- $0.40
- Cache read
- $0.005
Specifications
- Context window
- 400K
- Max output
- 128K
Benchmarks
Benchmarks
Sourced from LMArena.
Arena Score
1337.6
Overall Rank
#241 / 421
Votes
8,126
Confidence Interval
1330.7 - 1344.5
Category Scores
Coding
#235 / 416
1,664 votes
1385.0
Math
#219 / 404
493 votes
1347.7
Longer Query
#255 / 399
1,630 votes
1331.9
Creative Writing
#303 / 419
910 votes
1246.1
Instruction Following
#235 / 421
2,020 votes
1326.2
Hard Prompts
#236 / 421
3,777 votes
1355.9
Additional Categories21
Spanish
#171 / 303
235 votes
1383.4
German
#188 / 320
205 votes
1348.6
Polish
#198 / 244
571 votes
1336.8
Korean
#201 / 289
225 votes
1279.5
Japanese
#210 / 288
147 votes
1245.3
Chinese
#215 / 395
511 votes
1378.5
Industry Mathematical
#218 / 401
439 votes
1356.4
Industry Medicine And Healthcare
#222 / 390
470 votes
1366.3
Expert
#224 / 371
338 votes
1355.2
Non English
#224 / 421
4,546 votes
1327.7
Hard Prompts English
#235 / 419
1,693 votes
1370.4
Industry Business And Management And Financial Operations
#237 / 414
1,519 votes
1339.5
Exclude Ties
#240 / 421
5,882 votes
1292.9
Industry Software And It Services
#240 / 421
2,807 votes
1372.1
Industry Life And Physical And Social Science
#248 / 419
1,309 votes
1349.1
Multi Turn
#248 / 419
1,419 votes
1324.9
Russian
#251 / 385
601 votes
1311.7
English
#257 / 421
3,578 votes
1343.7
Industry Legal And Government
#266 / 393
470 votes
1328.6
Industry Entertainment And Sports And Media
#270 / 419
1,356 votes
1277.5
Industry Writing And Literature And Language
#282 / 420
1,726 votes
1288.2
Published 2026-11-09 · Matched as GPT-5 nano (High) · 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 12, 2026 | ||
Epoch Capabilities Index | 139.1 | #96 of 115 |
FrontierMath v2 (Tiers 1–3) | 20.0% | #75 of 83 |
GPQA Diamond | 69.4% | #92 of 119 |
SimpleQA Verified | 11.7% | #78 of 82 |
FrontierMath v2 (Tier 4) | 2.4% | #65 of 67 |
OTIS Mock AIME 2024–2025 | 81.1% | #81 of 121 |
Chess Puzzles | 22.0% | #58 of 82 |
Mystery Game Puzzles | 9.0% | #68 of 77 |
Creative Writing v3 | 703 | #94 of 96 |
Longform Writing | 28.0 | #94 of 97 |
Filtered and ranked to models on NanoGPT.
Providers
Auto routing is available for this model. Explicit provider selection is not available.
 Recommended | $0.05 | $0.40 | $0.005 | N/A | N/A |
