About
For tasks that demand low latency, GPT‑4.1 nano is the fastest and cheapest model in the GPT-4.1 series. It delivers exceptional performance at a small size with its 1 million...
Related Models
Pricing
Input
$0.10
per 1M tokens
Output
$0.40
per 1M tokens
Blended
$0.17
per 1M tokens
Cheaper than 72% of models. Median price is $0.54/1M tokens.
Cost Calculator
Tokens per day1M
100K100M
Daily
$0.17
Monthly
$5.25
vs. Similar Models
Mistral Large 2407Q:0.0
$3.00+1614%
Gemma 3 27B InstructQ:+0.1
$0.14-17%
NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning)Q:+0.1
$0.09-50%
NVIDIA Nemotron Nano 9B V2 (Non-reasoning)Q:+0.1
$0.09-51%
Performance
159
tokens/sec
Faster than 77% of models
0.54
seconds
Faster than 84% of models
0.54
seconds
Faster than 89% of models
Market Median
94 tok/s
70% faster
Median TTFT
1.10s
51% faster
Throughput/Dollar
911
tok/s per $/1M
Speed Comparison
OpenAI: GPT-5 Nano
160 tok/s+1%
GPT-5.4 nano (Non-Reasoning)
161 tok/s+1%
OpenAI: GPT-5.4
158 tok/s-1%
Context Window
1.0M
tokens
Larger than 89% of models
Max Output
33K
tokens
3% of context
Benchmarks
MMLU-Pro
65.7%
GPQA Diamond
51.2%
HLE
3.9%
LiveCodeBench
32.6%
SciCode
25.9%
TerminalBench Hard
3.8%
MATH-500
84.8%
AIME
23.7%
AIME 2025
24.0%
IFBench
32.0%
Long Context Recall
17.0%
Tau2
17.3%
Market AverageTop Score