GPT-5.4 is OpenAI's latest frontier model released in March 2026, unifying the Codex and GPT product lines into a single system. It features a 1M+ token context window, native computer-use capabilities, and industry-leading coding performance inherited from GPT-5.3-Codex. The model is significantly more token-efficient than GPT-5.2, and achieves state-of-the-art results on knowledge work benchmarks, matching or exceeding industry professionals in 83% of comparisons across 44 occupations. It excels at agentic coding, document understanding, tool use, and complex multi-step workflows.
OpenAI PlusOpenAI ProAPI|VisionReasoningWeb SearchFile|Proprietary Model
Knowledge Cutoff
2025-08-31
Input → Output Format
Context Memory
1.1MIN128KOUT
AI Performance Evaluation
Arena Overall Score
1481
±6As of 2026-04-23
Overall Rank
No.9
13,593 Votes
Arena by Ability
Hard Prompts
1503±7No.8
Expert Knowledge
1526±19No.5
Instruction Following
1481±10No.7
Conversation Memory
1497±12No.8
Creative
1448±14No.15
Coding
1532±11No.6
Math
1515±20🥈 No.2
Arena by Occupation
Creative Writing
1470±11No.7
Social Sciences
1479±13No.24
Media
1448±13No.14
Business
1477±12No.10
Healthcare
1475±20No.32
Legal
1471±19No.25
Software
1513±9No.9
Mathematics
1516±22🥉 No.3
Source:Arena Intelligence
Overall
AA Intelligence Index
57%↑18%
LiveBench
81%↑21%
ForecastBench
58%↓1%
Reasoning & Math
GPQA Diamond
92%↑11%
HLE
42%↑25%
LB Reasoning
88%↑28%
LB Math
94%↑21%
LB Data
79%↑30%
Coding
AA Coding Index
57%↑23%
LB Coding
78%↑4%
LB Agentic
70%↑27%
TAU2
87%↑14%
TerminalBench
58%↑26%
SciCode
57%↑16%
Language & Instructions
IFBench
74%↑17%
AA-LCR
74%↑12%
Hallucination (HHEM)
7.0%↓3%
Factual (HHEM)
93%↑3%
LB Language
83%↑11%
LB IF
70%↑24%
Output Speed
Standard Mode
155tok/s↑73
First Output 0.55s
Reasoning Mode
152tok/s↑64
First Output 7.32s
Multilingual Capabilities
MGSM 🇰🇷
94%
MGSM 🇯🇵
92%
KMMLU 🇰🇷
77%
JMMLU 🇯🇵
75%
