GP

GPT-5.4 nano

Multimodal modelGPT NanoGPT-5.4

GPT-5.4 nano

Release date: 2026-03-17Updated: 2026-06-15 07:18:17.329Knowledge cutoff: 2025-08-31525
Live demoGitHubHugging FaceCompare
Parameters
Not disclosed
Context length
400K
Chinese support
Supported
Reasoning ability

GPT-5.4 nano is an AI model published by OpenAI, released on 2026-03-17, for Multimodal model, and 400K context length, under the 不开源 license, with a 92.50 score on τ²-Bench - Telecom.

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 nano

Model basics

Reasoning traces
Supported
Thinking modes
Thinking Level · High (Default)Thinking Level · LowThinking Level · MediumThinking Level · Extra-High
Context length
400K tokens
Max output length
128K tokens
Model type
Multimodal model
Modality (in / out)
Text, Image → Text
Release date
2026-03-17
Model file size
No data
MoE architecture
No
Total params / Active params
No data / N/A
Knowledge cutoff
2025-08-31
GPT-5.4 nano

Open source & experience

Code license
不开源
Weights license
不开源
GitHub repo
GitHub link unavailable
Hugging Face
Hugging Face link unavailable
Live demo
No live demo
GPT-5.4 nano

Official resources

DataLearnerAI blog
No blog post yet
GPT-5.4 nano

API details

API speed
5/5
💡Default unit: $/1M tokens. If vendors use other units, follow their published pricing.
Standard
TypeConditionInputOutput
Text-$0.200/ 1M$1.25/ 1M
Batch
TypeConditionInputOutput
Text-$0.100/ 1M$0.625/ 1M
TextCache = 5 min$0.010/ 1M$0.625/ 1M
Cache PricingPrompt Cache
TypeTTLWriteRead
Text5m$0.020/ 1M$1.25/ 1M
GPT-5.4 nano

Benchmark Results

GPT-5.4 nano currently shows benchmark results led by LiveBench (38 / 115, score 70.13), Claw Bench (10 / 29, score 89.70), GPQA Diamond (63 / 179, score 82.80). 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

General Knowledge

8 evaluations
Benchmark / mode
Score
Rank/total
GPQA Diamond
Extra-High
82.80
63 / 179
LiveBench
Standard Mode
32.39
115 / 115
48.67
96 / 115
LiveBench
Medium
58.46
75 / 115
62.75
57 / 115
70.13
38 / 115
HLE
Extra-High
24.30
92 / 159
HLE
Extra-HighTools
37.70
57 / 159

Multimodal Understanding

2 evaluations
Benchmark / mode
Score
Rank/total
MMMU
Extra-High
66.10
26 / 28
MMMU
Extra-HighTools
69.50
24 / 28

Math and Reasoning

1 evaluations
Benchmark / mode
Score
Rank/total
6.30
35 / 80

Coding and Software Engineer

1 evaluations
Benchmark / mode
Score
Rank/total
SWE-Bench Pro - Public
Extra-HighTools
52.40
27 / 44

Agent Level Benchmark

1 evaluations
Benchmark / mode
Score
Rank/total
τ²-Bench - Telecom
Extra-HighTools
92.50
19 / 35

AI Agent - Tool Usage

3 evaluations
Benchmark / mode
Score
Rank/total
Terminal Bench 2.0
Extra-HighTools
46.30
40 / 46
OSWorld-Verified
Extra-HighTools
39
17 / 18
Tool Decathlon
Extra-HighTools
35.50
6 / 7

Claw-style Agent Evaluation

1 evaluations
Benchmark / mode
Score
Rank/total
Claw Bench
Thinking ModeTools
89.70
10 / 29

Compare with other models

No curated comparisons for this model yet.

Want a custom combination? Open the compare tool

GPT-5.4 nano

Publisher

GPT-5.4 nano

Model Overview

GPT-5.4 nano is an AI model published by OpenAI, released on 2026-03-17, for Multimodal model, and 400K context length, under the 不开源 license, with a 92.50 score on τ²-Bench - Telecom.

DataLearner on WeChat

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

DataLearner WeChat QR code