GPT-5 Nano is the smallest and fastest variant in the GPT-5 family, optimized for ultra-low latency and cost-sensitive environments. As the successor to GPT-4.1 Nano, it is purpose-built for developer tools, classification tasks, rapid interactions, and real-time applications where response speed takes priority over deep reasoning. Despite its compact size, it retains key instruction-following and safety features from the GPT-5 series.
API|VisionReasoningWeb SearchFile|Proprietary Model
Knowledge Cutoff
2024-05-31
Input → Output Format
Context Memory
400KIN128KOUT
AI Performance Evaluation
Arena Overall Score
1337
±7As of 2026-05-01
Overall Rank
No.180
8,278 Votes
Arena by Ability
Hard Prompts
1354±10No.175
Expert Knowledge
1352±34No.161
Instruction Following
1325±13No.173
Conversation Memory
1323±16No.190
Creative
1249±20No.239
Coding
1381±15No.178
Math
1344±27No.165
Arena by Occupation
Creative Writing
1290±14No.218
Social Sciences
1345±17No.188
Media
1278±16No.209
Business
1336±16No.177
Healthcare
1359±28No.166
Legal
1330±27No.202
Software
1371±11No.179
Mathematics
1353±29No.161
Source:Arena Intelligence
Overall
AA Intelligence Index
26%↓13%
LiveBench
48%↓13%
ForecastBench
57%↓2%
Reasoning & Math
AA Math Index
78%↑4%
GPQA Diamond
67%↓15%
HLE
7.6%↓10%
MMLU-Pro
77%↓4%
AIME 2025
78%↑4%
LB Reasoning
35%↓34%
LB Math
65%↓9%
LB Data
44%↓9%
Coding
AA Coding Index
23%↓14%
LiveCodeBench
76%↑11%
LB Coding
67%↓6%
LB Agentic
28%↓17%
TAU2
30%↓50%
TerminalBench
17%↓17%
SciCode
34%↓8%
Language & Instructions
IFBench
66%↑3%
AA-LCR
40%↓22%
Hallucination (HHEM)
11%↑0%
Factual (HHEM)
90%↑0%
LB Language
48%↓25%
LB IF
52%↑1%
Output Speed
Standard Mode
135tok/s↑58
First Output 0.73s
Reasoning Mode
141tok/s↑55
First Output 47.94s
