OpenAI: GPT-5.4 nano
gpt-5.4-nano- Input / Output-30%
- $0.20 / $1.25$0.14 / $0.875
- Context
- 400K
- Max output
- 128K
- Modalities
- VisionChat
- Capabilities
- Tool usePrompt Cache
About GPT-5.4 nano
GPT-5.4 nano is the smallest and cheapest model in the GPT-5.4 family. OpenAI targets simple high-volume tasks such as classification, data extraction, ranking, and subagents. It takes text and images, returns text, and supports tools including hosted shell and MCP. Reasoning effort runs from none to xhigh, and knowledge reaches August 2025.
Where it works well
- Aimed by OpenAI at classification, data extraction, ranking, and subagent work.
- Supports prompt caching and function calling at the smallest size in the family.
- Offers the same none to xhigh effort scale as bigger siblings.
- Knowledge reaches August 2025, later than GPT-5 nano.
When to choose another model
- It is the weakest GPT-5.4 model, so open-ended coding or reasoning belongs on mini or the full model.
- Image analysis is basic compared with larger siblings.
- 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-nano", "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.
Default spec
per 1M tokens- Official price
- Input $0.20 / Output $1.25 / Cache read $0.02
- Verified price
- Input $0.14 / Output $0.875 / Cache read $0.014
- Discount
- -30%
Cache read
- Official price
- $0.02
- Verified price
- $0.014
- 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 | |
|---|---|---|---|
| Default spec | Input $0.20 / Output $1.25 / Cache read $0.02 | Input $0.14 / Output $0.875 / Cache read $0.014 | -30% |
| Prompt cache pricing | |||
| Cache read | $0.02 | $0.014 | -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
- 99.9%
- 30-day requests
- 100+
- Last active
- yesterday
Data is based on aggregate user requests, excluding status checks.
Open in Console
Open GPT-5.4 nano in Console with a prompt ready to edit or send.
Help me try gpt-5.4-nano with a short message at /v1/responses. Show the reply, latency, and cost.
Use cases
Best for- Vision
Bulk extraction
Pull fields from large sets of emails, invoices, or logs into a fixed schema.
Ranking
Score candidates or documents against a rubric and return an ordered list in a fixed format.
Pre-routing
Decide which incoming messages need a larger model before spending on one, using a short classification prompt.
Lightweight subagent
Run simple file or text lookups under a coordinating model that handles planning and final review.
Prompt examples
Extract the company name, amount, and due date from this email into JSON.
Order these five documents by relevance to the query and explain the top choice in one line.
Is this message a refund request? Answer yes or no.
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 GPT-5.4 nano?
GPT-5.4 nano is OpenAI's cheapest model in the GPT-5.4 family, made for fast, simple, high-volume tasks that do not need deep reasoning. It is the right first model to try for simple, repetitive calls.
What is GPT-5.4 nano best at?
OpenAI lists classification, data extraction, ranking, and subagents as the target tasks, all of which favor speed and low cost per call. None of these need long chains of thought, so low effort is usually enough.
Does GPT-5.4 nano support tools?
Yes. Function calling, web search, file search, hosted shell, patch application, skills, and MCP are listed on its model page. Tool use works the same way as on larger siblings.
Should I choose GPT-5.4 nano or GPT-5 nano?
Pick 5.4 nano for the later knowledge cutoff and the wider effort range. Keep GPT-5 nano only if your prompts are already tuned and validated against it.
Can GPT-5.4 nano handle images?
It accepts image input and answers in text, which suits simple tagging or reading text from a picture, but not detailed visual analysis. Expect basic recognition rather than deep visual reasoning.
How much does GPT-5.4 nano cost?
On TokenLab, GPT-5.4 nano costs Input $0.14 / Output $0.875 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 nano?
GPT-5.4 nano accepts up to 400,000 tokens of context and returns up to 128,000 tokens in one response.
Which endpoint should GPT-5.4 nano use?
Use https://api.tokenlab.sh/v1/responses for GPT-5.4 nano. The request example below shows the matching code shape.
Which operations does GPT-5.4 nano support?
GPT-5.4 nano supports Text to text. Select an operation above to see its endpoint and request example.
Compare GPT-5.4 nano
Sources
Reviewed Oct 2, 2026
More from OpenAI