DataLearner logo
GP

GPT-5.4 Pro

Multimodal modelGPT ProGPT-5.4

GPT-5.4 Pro

Release date: 2026-03-05Updated: 2026-06-15 07:18:16.990Knowledge cutoff: 2025-081,431
Live demoGitHubHugging FaceCompare
Parameters
Not disclosed
Context length
1M
Chinese support
Supported
Reasoning ability

GPT-5.4 Pro is an AI model published by OpenAI, released on 2026-03-05, for Multimodal model, and 1M context length, under the 不开源 license, with a 94.50 score on ARC-AGI.

Data sourced primarily from official releases (GitHub, Hugging Face, papers), then benchmark leaderboards, then third-party evaluators. Learn about our data methodology

GPT-5.4 Pro

Model basics

Reasoning traces
Supported
Thinking modes
Thinking Level · High (Default)Thinking Level · Extra-High
Context length
1M tokens
Max output length
125K tokens
Model type
Multimodal model
Modality (in / out)
Text, Image → Text
Release date
2026-03-05
Model file size
No data
MoE architecture
No
Total params / Active params
No data / N/A
Knowledge cutoff
2025-08
GPT-5.4 Pro

Open source & experience

Code license
不开源
Weights license
不开源
GitHub repo
GitHub link unavailable
Hugging Face
Hugging Face link unavailable
GPT-5.4 Pro

Official resources

DataLearnerAI blog
No blog post yet
GPT-5.4 Pro

API details

API speed
2/5
💡Default unit: $/1M tokens. If vendors use other units, follow their published pricing.
Standard
TypeConditionInputOutput
TextContext <= 272K$30.00/ 1M$180.00/ 1M
TextContext > 272K$60.00/ 1M$270.00/ 1M
GPT-5.4 Pro

Benchmark Results

GPT-5.4 Pro currently shows benchmark results led by GPQA Diamond (2 / 180, score 94.40), HLE (3 / 163, score 58.70), BrowseComp (2 / 47, score 89.30). This page also consolidates core specs, context limits, and API pricing so you can evaluate the model from benchmark results and deployment constraints together.

Thinking
Tool usage
Internet

General Knowledge

5 evaluations
Benchmark / mode
Score
Rank/total
94.50
5 / 65
94.40
2 / 180
83.30
6 / 59
HLE
High
42.70
47 / 163
HLE
HighTools
58.70
3 / 163

Math and Reasoning

5 evaluations
Benchmark / mode
Score
Rank/total
50
3 / 60
FrontierMath
Extra-High
50
3 / 60
FrontierMath - Tier 4
Standard ModeToolsInternet
37.50
5 / 80
37.50
5 / 80

AI Agent - Information Search

1 evaluations
Benchmark / mode
Score
Rank/total
BrowseComp
HighTools
89.30
2 / 47

Productivity Knowledge

1 evaluations
Benchmark / mode
Score
Rank/total
GDPval-AA
HighTools
82
8 / 21

Compare with other models

GPT-5.4 Pro

Publisher

GPT-5.4 Pro

Model Overview

GPT-5.4 Pro is an AI model published by OpenAI, released on 2026-03-05, for Multimodal model, and 1M context length, under the 不开源 license, with a 94.50 score on ARC-AGI.

DataLearner on WeChat

Follow DataLearner on WeChat for AI model updates and research notes.

DataLearner WeChat QR code