GPT-4o mini
OpenAI's most cost-efficient small model. Cheaper and smarter than GPT-3.5 (the original ChatGPT), but less performant than gpt-4o.
- Vision
Pricing
Auto routing · per 1M tokens- Input
- $0.15
- Output
- $0.60
- Cache read
- $0.075
Specifications
- Context window
- 128K
- Max output
- 16.4K
Benchmarks
Benchmarks
Sourced from LMArena.
Arena Score
1317.6
Overall Rank
#270 / 421
Votes
68,705
Confidence Interval
1314.1 - 1321.2
Category Scores
Coding
#278 / 416
10,928 votes
1347.6
Math
#280 / 404
9,322 votes
1274.9
Longer Query
#260 / 399
8,989 votes
1326.0
Creative Writing
#251 / 419
10,486 votes
1293.8
Instruction Following
#277 / 421
26,707 votes
1292.9
Hard Prompts
#286 / 421
18,259 votes
1310.9
Additional Categories22
Japanese
#208 / 288
1,793 votes
1250.6
French
#223 / 301
649 votes
1337.8
German
#225 / 320
1,641 votes
1301.4
Polish
#230 / 243
65 votes
1277.3
Korean
#232 / 289
1,087 votes
1238.3
Spanish
#237 / 304
741 votes
1308.4
Industry Legal And Government
#238 / 393
3,866 votes
1346.5
Industry Writing And Literature And Language
#241 / 420
18,532 votes
1313.7
Russian
#254 / 385
8,268 votes
1309.4
Non English
#260 / 421
30,970 votes
1300.7
Industry Medicine And Healthcare
#262 / 391
3,439 votes
1334.0
Industry Entertainment And Sports And Media
#263 / 419
11,284 votes
1284.2
Chinese
#267 / 395
5,778 votes
1316.7
Multi Turn
#267 / 419
12,279 votes
1310.7
Industry Business And Management And Financial Operations
#268 / 414
7,765 votes
1310.5
Exclude Ties
#271 / 421
43,353 votes
1260.7
English
#278 / 421
37,735 votes
1328.2
Industry Life And Physical And Social Science
#278 / 419
11,641 votes
1326.3
Industry Mathematical
#278 / 402
7,966 votes
1285.4
Industry Software And It Services
#281 / 421
17,795 votes
1337.7
Expert
#284 / 371
3,550 votes
1284.6
Hard Prompts English
#290 / 419
11,355 votes
1319.1
Published 2026-11-11 · Matched as GPT-4o mini (2024-07-18) · 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 | 123.9 | #112 of 115 |
FrontierMath v2 (Tiers 1–3) | 0.7% | #82 of 83 |
GPQA Diamond | 37.7% | #117 of 119 |
SimpleQA Verified | 8.3% | #81 of 82 |
OTIS Mock AIME 2024–2025 | 6.9% | #116 of 121 |
Chess Puzzles | 3.0% | #80 of 82 |
Mystery Game Puzzles | 12.0% | #64 of 77 |
Creative Writing v3 | 871 | #89 of 96 |
Filtered and ranked to models on NanoGPT.
Providers
Auto routing is available for this model. Explicit provider selection is not available.
 Recommended | $0.15 | $0.60 | $0.075 | N/A | N/A |
