GPT-5.4 Nano is the most lightweight and cost-efficient model in the GPT-5.4 family, purpose-built for speed-critical, high-volume tasks. It supports text and image inputs and is optimized for low-latency use cases such as classification, data extraction, ranking, and sub-agent execution within larger systems. As a significant upgrade over GPT-5 Nano, it prioritizes responsiveness and efficiency, making it ideal for background pipelines, real-time systems, and distributed agent architectures where minimizing cost and latency is essential.
API|VisionReasoningWeb SearchFile|Proprietary Model
Knowledge Cutoff
2025-08-31
Input → Output Format
Context Memory
400KIN128KOUT
AI Performance Evaluation
Arena Overall Score
1406
±6As of 2026-05-01
Overall Rank
No.99
12,984 Votes
Arena by Ability
Hard Prompts
1423±7No.100
Expert Knowledge
1446±18No.73
Instruction Following
1388±10No.111
Conversation Memory
1416±12No.89
Creative
1337±13No.143
Coding
1467±11No.75
Math
1433±20No.54
Arena by Occupation
Creative Writing
1364±11No.128
Social Sciences
1427±13No.94
Media
1355±12No.119
Business
1412±12No.89
Healthcare
1432±20No.97
Legal
1407±19No.106
Software
1452±9No.86
Mathematics
1446±23No.48
Source:Arena Intelligence
Overall
AA Intelligence Index
38%↓1%
LiveBench
30%↓31%
ForecastBench
56%↓3%
Reasoning & Math
GPQA Diamond
76%↓6%
HLE
15%↓3%
LB Reasoning
17%↓52%
LB Math
36%↓38%
LB Data
39%↓14%
Coding
AA Coding Index
35%↓1%
LB Coding
62%↓11%
LB Agentic
27%↓18%
TAU2
53%↓28%
TerminalBench
33%↓1%
SciCode
38%↓3%
Language & Instructions
IFBench
64%↑1%
AA-LCR
57%↓5%
Hallucination (HHEM)
3.1%↓7%
Factual (HHEM)
97%↑7%
LB Language
29%↓44%
LB IF
16%↓34%
Output Speed
Standard Mode
197tok/s↑120
First Output 0.44s
Reasoning Mode
159tok/s↑72
First Output 2.63s
