GPT-5 mini
Multimodal
Zero-eval
#2HMMT 2025
#2FrontierMath
by OpenAI
About
GPT-5 mini is a multimodal language model developed by OpenAI. It achieves strong performance with an average score of 60.0% across 5 benchmarks. It excels particularly in AIME 2025 (91.1%), HMMT 2025 (87.8%), GPQA (82.3%). It supports a 528K token context window for handling large documents. The model is available through 2 API providers. As a multimodal model, it can process and understand text, images, and other input formats seamlessly. Released in 2025, it represents OpenAI's latest advancement in AI technology.
Pricing Range
Input (per 1M)$0.25 -$0.25
Output (per 1M)$2.00 -$2.00
Providers2
Timeline
AnnouncedAug 7, 2025
ReleasedAug 7, 2025
Knowledge CutoffMay 30, 2024
Specifications
Capabilities
Multimodal
License & Family
License
Proprietary
Benchmark Performance Overview
Performance metrics and category breakdown
Overall Performance
5 benchmarks
Average Score
60.0%
Best Score
91.1%
High Performers (80%+)
3Performance Metrics
Max Context Window
528.0KAvg Throughput
200.0 tok/sAvg Latency
1msTop Categories
general
69.5%
math
22.1%
Benchmark Performance
Top benchmark scores with normalized values (0-100%)
Ranking Across Benchmarks
Position relative to other models on each benchmark
AIME 2025
Rank #6 of 36
#3Grok-4
91.7%
#4o4-mini
92.7%
#5Grok-3
93.3%
#6GPT-5 mini
91.1%
#7Grok-3 Mini
90.8%
#8Gemini 2.5 Pro Preview 06-05
88.0%
#9DeepSeek-R1-0528
87.5%
HMMT 2025
Rank #2 of 7
#1GPT-5
93.3%
#2GPT-5 mini
87.8%
#3DeepSeek-R1-0528
79.4%
#4GPT-5 nano
75.6%
#5Kimi K2 Instruct
38.8%
GPQA
Rank #11 of 115
#8Gemini 2.5 Flash
82.8%
#9Gemini 2.5 Pro
83.0%
#10o3
83.3%
#11GPT-5 mini
82.3%
#12o4-mini
81.4%
#13DeepSeek-R1-0528
81.0%
#14Claude Opus 4
79.6%
FrontierMath
Rank #2 of 6
#1GPT-5
26.3%
#2GPT-5 mini
22.1%
#3o3
15.8%
#4GPT-5 nano
9.6%
#5o3-mini
9.2%
Humanity's Last Exam
Rank #8 of 16
#5DeepSeek-R1-0528
17.7%
#6Gemini 2.5 Pro
17.8%
#7o3
20.2%
#8GPT-5 mini
16.7%
#9o4-mini
14.7%
#10Gemini 2.5 Flash
11.0%
#11Magistral Medium
9.0%
All Benchmark Results for GPT-5 mini
Complete list of benchmark scores with detailed information
AIME 2025 AIME 2025 benchmark | general | text | 0.91 | 91.1% | Self-reported |
HMMT 2025 HMMT 2025 benchmark | general | text | 0.88 | 87.8% | Self-reported |
GPQA GPQA benchmark | general | text | 0.82 | 82.3% | Self-reported |
FrontierMath FrontierMath benchmark | math | text | 0.22 | 22.1% | Self-reported |
Humanity's Last Exam Humanity's Last Exam benchmark | general | text | 0.17 | 16.7% | Self-reported |
Resources