GPT model for general reasoning, writing, coding, and tool-assisted tasks
Specification
most-agreed values
Context
66K
Max output
8K
Released
2025-04-15
Knowledge cutoff
2024-04-30
Retires
—
Open weights
yes
Input
text, pdf
Output
text
Price
US dollars per million tokens · most-agreed
Input
$0.493
Output
$0.493
Cache read
$0.2465
Quality
2 benchmarks · 1 source
Composite
21st
percentile of 334 scored models
Rank
265 / 334
±0.06 sd
Evidence
2 × 1
too little evidence to place with confidence
Effort range
—
one configuration only
reasoning22nd1/4 bench
math19th1/7 bench
Every score
one row per source and configuration — nothing averaged away
| Benchmark | Score | Configuration | Source | Run |
|---|---|---|---|---|
| GPQA diamond | 43.4 ±2.3 | default | Epoch AI | 2025-01-27 |
| MATH level 5 | 25.7 ±0.9 | default | Epoch AI | 2025-01-27 |
The leading figure for a benchmark is the median across its configurations, so one heroic high-effort run cannot set the number.
Available from 6 hosts
More from Microsoft
most-hosted first