GPT 5 Mini
A lightweight version of GPT-5 for cost-sensitive applications. Balances advanced capabilities with efficient resource usage.
- Reasoning
- Vision
Added Aug 7, 2025
Pricing
Auto routing · per 1M tokens- Input
- $0.25
- Output
- $2.00
- Cache read
- $0.025
Specifications
- Context window
- 400K
- Max output
- 128K
Benchmarks
Benchmarks
Sourced from LMArena.
Arena Score
1389.9
Overall Rank
#182 / 421
Votes
26,718
Confidence Interval
1385.2 - 1394.5
Category Scores
Coding
#189 / 416
5,497 votes
1431.8
Math
#154 / 404
1,445 votes
1407.3
Longer Query
#205 / 399
5,883 votes
1370.5
Creative Writing
#216 / 419
3,514 votes
1323.2
Instruction Following
#191 / 421
6,990 votes
1372.3
Hard Prompts
#188 / 421
12,817 votes
1404.0
Additional Categories22
Japanese
#129 / 288
459 votes
1364.7
Polish
#140 / 244
1,562 votes
1399.2
German
#151 / 320
590 votes
1385.4
French
#162 / 301
414 votes
1409.4
Industry Mathematical
#168 / 401
1,281 votes
1408.9
Korean
#175 / 289
655 votes
1322.5
Non English
#175 / 421
15,260 votes
1377.8
Russian
#178 / 385
1,868 votes
1378.3
Expert
#179 / 371
1,217 votes
1403.6
Industry Medicine And Healthcare
#180 / 390
1,435 votes
1407.5
Industry Legal And Government
#181 / 393
1,690 votes
1396.6
Exclude Ties
#183 / 421
19,453 votes
1366.4
Industry Entertainment And Sports And Media
#186 / 419
4,832 votes
1348.3
Spanish
#186 / 303
853 votes
1372.7
Chinese
#187 / 395
1,947 votes
1403.4
Industry Business And Management And Financial Operations
#187 / 414
4,821 votes
1381.7
English
#189 / 421
11,417 votes
1397.9
Hard Prompts English
#192 / 419
5,572 votes
1409.5
Industry Life And Physical And Social Science
#192 / 419
4,190 votes
1397.6
Industry Software And It Services
#192 / 421
8,887 votes
1416.2
Multi Turn
#198 / 419
4,380 votes
1372.6
Industry Writing And Literature And Language
#200 / 420
6,127 votes
1350.9
Published 2026-11-09 · Matched as GPT-5 mini (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 | 146.1 | #80 of 115 |
FrontierMath v2 (Tiers 1–3) | 46.7% | #56 of 83 |
SWE-bench Verified | 64.7% | #23 of 28 |
GPQA Diamond | 75.0% | #89 of 119 |
SimpleQA Verified | 21.6% | #71 of 82 |
FrontierMath v2 (Tier 4) | 12.2% | #57 of 67 |
OTIS Mock AIME 2024–2025 | 86.7% | #65 of 121 |
Chess Puzzles | 24.0% | #54 of 82 |
Mystery Game Puzzles | 10.0% | #67 of 77 |
Creative Writing v3 | 1314 | #72 of 96 |
Longform Writing | 56.2 | #63 of 97 |
Filtered and ranked to models on NanoGPT.
Providers
Auto routing is available for this model. Explicit provider selection is not available.
 Recommended | $0.25 | $2.00 | $0.025 | N/A | N/A |
