GPT-5 mini
GPT-5 mini is an API-only (closed weights) language model from OpenAI, released August 7, 2025. The context window is 391K tokens. On the Epoch Capabilities Index it scores 146.1, rank 91 of 256 rated models. API price: $0.25 input / $2 output per million tokens.
Check it against your own PC →
Lab
OpenAI
Released
August 7, 2025
Weights
API-only (closed weights)
Parameters
–
Context
391K
Price / M tokens
$0.25 / $2 in / out
Epoch ECI
146.1 #91
Strengths by category
ModelTape's 0–100 fit on Epoch AI benchmark results.
| Category | Score | Based on |
|---|---|---|
| Reasoning | 13.7 | Chess Puzzles, Mystery Game Puzzles, Humanity's Last Exam |
| Math | 37.6 | ProofBench, MATH level 5, OTIS Mock AIME 2024-2025, FrontierMath-Tiers-1-3-v2-Private, FrontierMath-Tier-4-v2-Private |
| Coding | 40.1 | Terminal Bench, SWE-bench Verified |
| Knowledge | 57.3 | GPQA diamond, SimpleQA Verified |
Benchmark results
| Benchmark (Epoch AI) | Score |
|---|---|
| MATH level 5 | 97.8% |
| OTIS Mock AIME | 86.7% |
| GPQA Diamond | 69.4% |
| SWE-bench Verified | 64.7% |
| FrontierMath (Tiers 1–3) | 49.1% |
| Terminal-Bench | 34.8% |
| SimpleQA Verified | 21.6% |
| Chess Puzzles | 20.0% |
| Humanity's Last Exam | 19.4% |
| FrontierMath-Tier-4-v2-Private | 12.2% |
| ProofBench | 9.0% |
| Mystery Game Puzzles | 0.9% |
Links
Announcement · Epoch AI benchmarks
Models with a similar score
GPT-5 · Grok 4.3 Beta · Qwen 3.6 Max (Preview) · GPT-5.4 Mini · Qwen 3.6 Plus · Kimi K2.5 · Inkling · Qwen3.5 397B-A17B
More from OpenAI
GPT-6.1 Terra · GPT-6.1 Sol · GPT-6 Sol · GPT-6 Luna · GPT-6 Astra · GPT-5.6 Cyber · Daybreak Red · Daybreak Blue