GPT-5.4 nano
Current
- Context window
- 400.000 tokens
- Max output
- 16.384 tokens
- Price input / 1M tokens
- $0.20
- Price output / 1M tokens
- $1.25
- Price cache-read / 1M
- $0.02
Capabilities
- Vision
- Tool use
Benchmark scores
- MMLU
- 82.0
- SWE-bench Pro
- 52.4
- OSWorld-Verified
- 39.0
- Terminal-Bench 2.0
- 46.3
Notes
Smallest GPT-5.4 variant, for high-volume and simple tasks. Source: openai.com/api/pricing.