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
1405
±6As of 2026-04-23
Overall Rank
No.93
10,607 Votes
Arena by Ability
Hard Prompts
1420±8No.100
Expert Knowledge
1449±20No.58
Instruction Following
1385±11No.106
Conversation Memory
1414±14No.81
Creative
1336±15No.136
Coding
1460±12No.78
Math
1428±22No.56
Arena by Occupation
Creative Writing
1362±12No.122
Social Sciences
1421±15No.92
Media
1351±14No.116
Business
1412±13No.81
Healthcare
1428±23No.96
Legal
1407±20No.102
Software
1447±10No.82
Mathematics
1442±26No.47
Source:Arena Intelligence
Overall
AA Intelligence Index
38%↑0%
LiveBench
30%↓30%
ForecastBench
55%↓4%
Reasoning & Math
GPQA Diamond
76%↓5%
HLE
15%↓2%
LB Reasoning
17%↓42%
LB Math
36%↓38%
LB Data
39%↓10%
Coding
AA Coding Index
35%↑1%
LB Coding
62%↓12%
LB Agentic
27%↓16%
TAU2
53%↓21%
TerminalBench
33%↑2%
SciCode
38%↓2%
Language & Instructions
IFBench
64%↑8%
AA-LCR
57%↓4%
Hallucination (HHEM)
3.1%↓7%
Factual (HHEM)
97%↑7%
LB Language
29%↓43%
LB IF
16%↓30%
Output Speed
Standard Mode
197tok/s↑115
First Output 0.44s
Reasoning Mode
152tok/s↑64
First Output 3.40s
