GPT-5.4 API

A high-performance large language model engineered for advanced reasoning, multimodal understanding, long-context processing, and real-world coding tasks.

Modalities: TextText
In / Out price: $2.5/$15per 1M
Context window: 1.05M tokens
Released: Mar 16, 2026

GPT-5.4 overview

GPT-5.4 specs

Model ID: openai/gpt-5-4
Developer: OpenAI
Context window: 1,050,000 tokens
Max output: 128,000 tokens
Input: Text
Output: Text
API: OpenAI-compatible, base URL https://api.aimlapi.com/v1

What exactly is GPT-5.4 API?

Unlike earlier GPT models that optimized primarily for fluency, GPT-5.4 pushes the frontier on structured reasoning and multi-step problem solving. It's not just generating plausible text, it's working through problems the way a thoughtful analyst would, weighing evidence, catching contradictions, and arriving at well-grounded conclusions.

The model is designed to operate at scale across enterprise workflows, developer toolchains, and research environments, handling everything from a quick summarization task to a multi-document analysis spanning thousands of tokens in a single pass.

Smarter, Faster, More Connected

GPT-5.4 isn’t just an upgrade, it’s a new generation of cognitive AI performance. Leveraging enhanced transformer architecture and a refined training pipeline, it delivers exceptional fluency, logical coherence, and contextual memory across text, image, and structured data inputs.

Key Highlights

Extended Context Window: Processes and remembers over a million tokens, ideal for handling entire books, technical documentation, or extensive chat histories.‍
Unmatched Reasoning Power: Solves complex logic chains, mathematical proofs, and programming tasks with near-human accuracy.‍
Adaptive Multimodality: Understands and generates text, code, images, and audio descriptions seamlessly.‍
Proactive Learning: Continually refines responses based on nuanced patterns in user input for a more responsive, human-like dialogue.

Technical specifications

For developers and engineers evaluating GPT-5.4 for integration, here's a closer look at what the model brings to the table technically.

Model family: GPT-5 series (OpenAI)
Architecture: Transformer-based autoregressive language model with RLHF fine-tuning
Modalities: Text + image input; text output
‍Context window: 128,000+ tokens, supports book-length documents in a single pass
Output length: Extended output with coherence maintained across thousands of tokens
Supported languages: 100+ natural languages; strong multilingual performance
Primary use cases: Reasoning, coding, analysis, document QA, conversational AI, agentic tasks

Long-Context Intelligence

GPT-5.4 introduces persistent context tracking, preserving meaning across massive documents and multi-turn conversations. This capability empowers professional applications such as technical manuals, legal analyses, or serialization of creative writing without loss of coherence or tone.

Example: GPT-5.4 can analyze a 200-page research report, summarize its core insights, and generate concise recommendations tailored to different audiences, from scientists to investors.

Multimodal Reasoning Engine

Beyond text, GPT-5.4 handles visual elements, numeric data, and symbolic logic natively. It can describe images, interpret charts, or generate structured JSON outputs in a single prompt. This integrated reasoning pipeline supports advanced workflows in AI-driven design, synthetic biology, and data visualization.

Use cases

GPT-5.4 is documented as suited to computer-use workflows, including operating computers and carrying out complex workflows across applications. It is also suited to coding tasks, especially writing code to operate computers with libraries like Playwright and issuing mouse and keyboard commands from screenshots.

It is documented as suited to deep web research, particularly for highly specific queries, while maintaining context over longer thinking. It is also suited to professional work involving spreadsheets, presentations, and documents, where it helps get complex work done accurately and efficiently.

Capabilities

Function calling
Streaming
Structured outputs
Reasoning
Web search

GPT-5.4 benchmarks by Artificial Analysis

See how GPT-5.4 compares in generation speed. Data retrieved Nov 3, 2026.

Speed#103 / 219108Output tokens / second Humanity's Last Exam#30 / 45843.7%Score GPQA Diamond#23 / 43992%Score AA-LCR#33 / 39182%Score

Speed

Output tokens / second

Benchmark results

Compare 17 models available on AI/ML API across 5 independent benchmarks.

Humanity's Last Exam

Expert-written questions across academic subjects.

GPT-5.4 API pricing

GPT-5.4 pricing in US dollars
Usage	Price	Unit
Input	$2.5	1M tokens
Output	$15	1M tokens
Cached input	$0.25	1M tokens

Prices from the AI/ML API price list, checked on Nov 4, 2026.

GPT-5.4 vs other models: token prices

AI/ML API input and output rates for 5 models, per million tokens.

AI/ML API model prices in USD per million tokens
Model	Input	Output
GPT-5.4	$2.5	$15
GPT-6 Luna	$0.1	$0.5
GPT-6.1 Terra	$1	$5
GPT-6.1 Sol	$2	$10
GPT-6.1 Terra Pro	$2.12	$10.58

How to use the GPT-5.4 API

View docs

Build with your agent

Give your agent the model details, code and documentation.

1Get your API key

Create API key

export AIMLAPI_KEY="8f2b4c7d9a1e6b3f…"

2Make your first request

curl 'https://api.aimlapi.com/v1/chat/completions' \
  -H "Authorization: Bearer $AIMLAPI_KEY" \
  -H 'Content-Type: application/json' \
  -d '{
  "model": "openai/gpt-5-4",
  "messages": [
    {
      "role": "user",
      "content": "Write a haiku about the ocean."
    }
  ]
}'

import os
import requests

headers = {
    "Authorization": f"Bearer {os.environ['AIMLAPI_KEY']}",
    "Content-Type": "application/json",
}
payload = {
    "model": "openai/gpt-5-4",
    "messages": [
        {
            "role": "user",
            "content": "Write a haiku about the ocean."
        }
    ]
}

response = requests.post("https://api.aimlapi.com/v1/chat/completions", headers=headers, json=payload, timeout=120)
response.raise_for_status()
print(response.json())

const response = await fetch("https://api.aimlapi.com/v1/chat/completions", {
  method: "POST",
  headers: {
    Authorization: `Bearer ${process.env.AIMLAPI_KEY}`,
    "Content-Type": "application/json",
  },
  body: JSON.stringify({
    "model": "openai/gpt-5-4",
    "messages": [
      {
        "role": "user",
        "content": "Write a haiku about the ocean."
      }
    ]
  }),
});

if (!response.ok) throw new Error(`HTTP ${response.status}`);
console.log(await response.json());

All posts

GPT-5.4 FAQ

What is the context window of GPT-5.4?

GPT-5.4 has a 1,050,000 tokens context window and can return up to 128,000 tokens.

What inputs and outputs does GPT-5.4 support?

GPT-5.4 takes image, text as input and returns text.

What is the model ID for GPT-5.4?

Use openai/gpt-5-4 as the model id. Requests go to https://api.aimlapi.com/v1/chat/completions.

How much does GPT-5.4 cost?

GPT-5.4 costs $2.5 input and $15 output per 1M tokens on AI/ML API, billed as you go with no subscription.

Does GPT-5.4 support streaming?

Yes, GPT-5.4 can stream responses as they are generated.

Can GPT-5.4 read images?

Yes, GPT-5.4 accepts image input alongside text.

Who created GPT-5.4?

GPT-5.4 was built by OpenAI.

Start building with GPT-5.4

Get API key

GPT-5.4 API

openai/gpt-5-4

Build with your agent

Give your agent the model details, code and documentation.

1Get your API key

Create API key

export AIMLAPI_KEY="8f2b4c7d9a1e6b3f…"

2Make your first request

curl 'https://api.aimlapi.com/v1/chat/completions' \
  -H "Authorization: Bearer $AIMLAPI_KEY" \
  -H 'Content-Type: application/json' \
  -d '{
  "model": "openai/gpt-5-4",
  "messages": [
    {
      "role": "user",
      "content": "Write a haiku about the ocean."
    }
  ]
}'

import os
import requests

headers = {
    "Authorization": f"Bearer {os.environ['AIMLAPI_KEY']}",
    "Content-Type": "application/json",
}
payload = {
    "model": "openai/gpt-5-4",
    "messages": [
        {
            "role": "user",
            "content": "Write a haiku about the ocean."
        }
    ]
}

response = requests.post("https://api.aimlapi.com/v1/chat/completions", headers=headers, json=payload, timeout=120)
response.raise_for_status()
print(response.json())

const response = await fetch("https://api.aimlapi.com/v1/chat/completions", {
  method: "POST",
  headers: {
    Authorization: `Bearer ${process.env.AIMLAPI_KEY}`,
    "Content-Type": "application/json",
  },
  body: JSON.stringify({
    "model": "openai/gpt-5-4",
    "messages": [
      {
        "role": "user",
        "content": "Write a haiku about the ocean."
      }
    ]
  }),
});

if (!response.ok) throw new Error(`HTTP ${response.status}`);
console.log(await response.json());

Docs for AI agents Plain-text index · llms.txt API documentation API reference and examples

Talk to our sales team

Thanks for reaching out

GPT-5.4 API

GPT-5.4 overview

GPT-5.4 specs

What exactly is GPT-5.4 API?

Smarter, Faster, More Connected

Key Highlights

Technical specifications

Long-Context Intelligence

Multimodal Reasoning Engine

Use cases

Capabilities

GPT-5.4 benchmarks by Artificial Analysis

Speed

Benchmark results

Humanity's Last Exam

GPQA Diamond

AA-LCR

IFBench

τ²-Bench Telecom

GPT-5.4 API pricing

GPT-5.4 vs other models: token prices

How to use the GPT-5.4 API

Build with your agent

1Get your API key

2Make your first request

Related articles

GPT-5.4 FAQ

Start building with GPT-5.4

GPT-5.4 API

Build with your agent

1Get your API key

2Make your first request

GPT-5.4 API

GPT-5.4 overview

GPT-5.4 specs

What exactly is GPT-5.4 API?

Smarter, Faster, More Connected

Key Highlights

Technical specifications

Long-Context Intelligence

Multimodal Reasoning Engine

Use cases

Capabilities

Speed

Benchmark results

Humanity's Last Exam

GPQA Diamond

AA-LCR

IFBench

τ²-Bench Telecom

GPT-5.4 API pricing

GPT-5.4 vs other models: token prices

How to use the GPT-5.4 API

Build with your agent

1Get your API key

2Make your first request

Models similar to GPT-5.4

GPT-6.1 Terra

GPT-6.1 Terra Pro

GPT-6.1 Sol

Qwen3.8 2.4T A95B

Pareto

Command A

Related articles

Best AI Models for Agentic Workflows and Tool Use in 2026

Best LLMs for Long-Context & Multimodal Tasks in 2026

GPT-5.4 FAQ

Start building with GPT-5.4