OpenAI: GPT-5.4
gpt-5.4- Input / Output-30%
- $2.50 / $15.00$1.75 / $10.50
- Context
- 1.05M
- Max output
- 128K
- Modalities
- VisionChat
- Capabilities
- Tool usePrompt CacheReasoning
About GPT-5.4
GPT-5.4 is an OpenAI flagship for complex professional work, with a window large enough for whole codebases and document sets. It takes text and images, offers reasoning effort from none to xhigh with none as the default, and adds computer use to its tool list. Its knowledge runs through August 2025.
Where it works well
- Computer use is a listed tool, so it can operate software through screenshots.
- Reasoning effort runs none to xhigh with none as the default, which keeps simple calls fast.
- Supports function calling, structured JSON output and prompt caching alongside images.
- Its large context window suits long codebases and document collections.
When to choose another model
- OpenAI's own page now steers new work to later releases such as GPT-5.5.
- Defaulting to no reasoning means hard tasks need effort raised explicitly.
- It returns text only, never images or audio.
Getting started
Create API key
Create a key in Console, then use it with every model on the platform.
Send your first request
Copy the example for your language and run it against the endpoint.
Text to textResponses APIPOST/v1/responsesAPI format:curl https://api.tokenlab.sh/v1/responses \ -H "Content-Type: application/json" \ -H "Authorization: Bearer sk-xxx" \ -d '{ "model": "gpt-5.4", "input": "Hello!" }'
Pricing
The Verified price applies to Verified, which costs less on most models. The Official price is the model maker's published price and applies to the more reliable Official route. Auto bills the route that completes the request.
Input tokens <= 272K
per 1M tokens- Official price
- Input $2.50 / Output $15.00 / Cache read $0.25
- Verified price
- Input $1.75 / Output $10.50 / Cache read $0.175
- Discount
- -30%
Input tokens 272K-1.05M
per 1M tokens- Official price
- Input $5.00 / Output $22.50 / Cache read $0.50
- Verified price
- Input $3.50 / Output $15.75 / Cache read $0.35
- Discount
- -30%
Cache read
- Official price
- $0.25
- Verified price
- $0.175
- Discount
- -30%
Charged per call, only when the model uses the tool.
Web search
- Official price
- $0.01/search
- Verified price
- $0.007/search
- Discount
- -30%
| Official priceper 1M tokens | Verified priceper 1M tokens | Discount | |
|---|---|---|---|
| Input tokens <= 272K | Input $2.50 / Output $15.00 / Cache read $0.25 | Input $1.75 / Output $10.50 / Cache read $0.175 | -30% |
| Input tokens 272K-1.05M | Input $5.00 / Output $22.50 / Cache read $0.50 | Input $3.50 / Output $15.75 / Cache read $0.35 | -30% |
| Prompt cache pricing | |||
| Cache read | $0.25 | $0.175 | -30% |
| Tool feesCharged per call, only when the model uses the tool. | |||
| Web search | $0.01/search | $0.007/search | -30% |
Usage & activity
Success rate is the share of requests that completed. Latency is how long a full response takes; P95 means 95% of requests finished within that time.
Usage & availability
Last 24 hours- Requests
- Success rate
- P95 latency
- Total tokens
- 30-day success rate
- 99.5%
- 30-day requests
- 100+
- Last active
- 5 hours ago
Data is based on aggregate user requests, excluding status checks.
Open in Console
Open GPT-5.4 in Console with a prompt ready to edit or send.
Help me try gpt-5.4 with a short message at /v1/responses. Show the reply, latency, and cost.
Use cases
Best for- Reasoning
- Vision
Computer-use agents
Drive a browser or desktop app from screenshots for form filling and data entry.
Large codebase tasks
Load many source files at once and make coordinated changes across modules without losing track of how they connect.
Professional analysis
Review long filings, reports, or contracts and produce structured findings that point back to the passages used.
Tool-heavy assistants
Combine web search, file search, MCP servers, and your own functions in one conversation without switching models.
Prompt examples
Open the vendor portal, download last month's invoices, and list any total that differs from our records.
Read these twelve source files and explain how a request flows from the controller to the database.
Summarize the risk factors that changed between these two annual reports.
This model has conditional pricing. Monthly token totals alone cannot produce a reliable estimate; use the detailed pricing for the request specification, cache, and applicable time window.
FAQ
What is OpenAI's GPT-5.4 model?
GPT-5.4 is OpenAI's model for complex professional work, with computer use, a long context window, and reasoning effort from none to xhigh. It also handles images alongside text in the same request.
Does GPT-5.4 reason by default?
No. OpenAI lists none as the default effort, so a plain call skips the reasoning phase. Set low through xhigh when a task needs deliberate step-by-step thinking.
Can GPT-5.4 control a computer?
Computer use is listed among its supported tools, which lets it act on screens through screenshots, clicks, and typing in a loop, so it can fill forms and move between applications.
What does GPT-5.4 not support?
OpenAI's model page lists it without Realtime sessions or fine-tuning. It does take function calling, structured outputs, and prompt caching, so existing OpenAI-style client code works with only the model name changed.
How does GPT-5.4 relate to GPT-5.4 mini and nano?
Mini and nano are the smaller, cheaper siblings for subagents and simple tasks, with a smaller context window. GPT-5.4 is the full model for complex work.
How much does GPT-5.4 cost?
On TokenLab, GPT-5.4 costs Input $1.75 / Output $10.50 per 1M tokens. The pricing table above shows the full breakdown. Rates depend on the billing unit, specification, and usage. Compare matching conditions in the model's detailed pricing; a single rate does not determine the total cost.
What are the context window and output limit of GPT-5.4?
GPT-5.4 accepts up to 1,050,000 tokens of context and returns up to 128,000 tokens in one response.
Which endpoint should GPT-5.4 use?
Use https://api.tokenlab.sh/v1/responses for GPT-5.4. The request example below shows the matching code shape.
Which operations does GPT-5.4 support?
GPT-5.4 supports Text to text. Select an operation above to see its endpoint and request example.
Compare GPT-5.4
Sources
Reviewed Oct 2, 2026
More from OpenAI