GPT-4.1 nano

GPT-4.1 nano is OpenAI's fastest and cheapest model available in the GPT-4.1 family. It delivers exceptional performance at a small size with its 1 million token context window.

MMLU

80.1%

i
IFEval

74.5%

i
CharXiv-D

73.9%

i
MMMLU

66.9%

i
Multi-IF

57.2%

i
MathVista

56.2%

i
MMMU

55.4%

i
GPQA

50.3%

i
COLLIE

42.5%

i
CharXiv-R

40.5%

i
OpenAI-MRCR: 2 needle 128k

36.6%

i
Internal API instruction following (hard)

31.6%

i
MultiChallenge (o3-mini grader)

31.1%

i
AIME 2024

29.4%

i
Graphwalks BFS <128k

25.0%

i
TAU-bench Retail

22.6%

i
Multi-Challenge

15.0%

i
TAU-bench Airline

14.0%

i
OpenAI-MRCR: 2 needle 1M

12.0%

i
Aider-Polyglot

9.8%

i
Graphwalks parents <128k

9.4%

i
Aider-Polyglot Edit

6.2%

i
ComplexFuncBench

5.7%

i
Graphwalks parents >128k

5.6%

i
Graphwalks BFS >128k

2.9%

i

Pricing, uptime, and speed via OpenRouter — updated Jul 17, 2026, 04:19 AM.

Provider	Status	Input	Output	Limits	Uptime	Speed	Notes
OpenAI	available	$0.10/Mtok cache $0.02/Mtok	$0.40/Mtok	1.0M tokens context 33K tokens max output	99.8% 5m 99.6%	412 ms p50 TTFT 67 tok/s p50	$0.01/web search
Azure	available	$0.11/Mtok cache $0.03/Mtok	$0.44/Mtok	1.0M tokens context 33K tokens max output	—	—	cache $0.01/web search