OpenAI: GPT-5.4 mini
gpt-5.4-mini- Input / Output-30%
- $0.75 / $4.50$0.525 / $3.15
- Context
- 400K
- Max output
- 128K
- Modalities
- VisionChat
- Capabilities
- Tool usePrompt CacheReasoning
About GPT-5.4 mini
GPT-5.4 mini is OpenAI's strongest small model for coding, computer use, and subagents. It brings much of GPT-5.4's behavior to a faster, cheaper model for high-volume work. It accepts text and images, supports computer use and MCP, and offers reasoning effort from none to xhigh with a knowledge cutoff of August 2025.
Where it works well
- Built for subagent roles, where a larger model plans and small models run focused parallel tasks.
- Computer use is supported, unusual for a small model.
- Reasoning effort runs none to xhigh, defaulting to none for speed.
- Function calling and prompt caching are supported for high-volume tool use.
When to choose another model
- Its context window is smaller than the full GPT-5.4.
- Complex planning and ambiguous tasks are better left to GPT-5.4 or a Pro tier.
- 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-mini", "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 $0.75 / Output $4.50 / Cache read $0.075
- Verified price
- Input $0.525 / Output $3.15 / Cache read $0.0525
- Discount
- -30%
Cache read
- Official price
- $0.075
- Verified price
- $0.0525
- 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 $0.75 / Output $4.50 / Cache read $0.075 | Input $0.525 / Output $3.15 / Cache read $0.0525 | -30% |
| Prompt cache pricing | |||
| Cache read | $0.075 | $0.0525 | -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
- 99.1%
- 30-day requests
- 4K+
- Last active
- 1 minute ago
Data is based on aggregate user requests, excluding status checks.
Open in Console
Open GPT-5.4 mini in Console with a prompt ready to edit or send.
Help me try gpt-5.4-mini with a short message at /v1/responses. Show the reply, latency, and cost.
Use cases
Best for- Reasoning
- Vision
Coding subagents
Search a codebase, read files, and report findings back to a coordinating model.
Screenshot-driven automation
Handle repetitive on-screen steps through screenshots where the cost per action matters more than peak accuracy.
High-volume support
Answer routine customer questions with tool lookups at low latency, and escalate unusual cases to a larger model.
Multimodal triage
Read attached images and decide which team or follow-up workflow each case belongs to.
Prompt examples
Find every place this config key is read in the repo and list file and line for each.
Look at this dashboard screenshot and tell me which service shows an error spike.
Draft a reply to this ticket using the order lookup tool and keep it under 100 words.
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 mini?
GPT-5.4 mini is the small model of the GPT-5.4 family. OpenAI positions it as its strongest mini model for coding, computer use, and subagents. It is sized for high-volume, latency-sensitive calls.
What is a subagent model?
A subagent handles a narrow task delegated by a larger planning model. Mini suits searching code, reviewing files, and running parallel lookups for a coordinating model.
How is GPT-5.4 mini different from GPT-5.4 nano?
Mini is the more capable of the two and supports computer use in its description. Nano targets simple classification, extraction, and ranking at the lowest cost.
Does GPT-5.4 mini accept images?
Yes. Image input is supported alongside text, and the output is text only. That lets it read screenshots and photographed documents for triage or extraction.
Can GPT-5.4 mini reason?
Yes. Effort levels none, low, medium, high, and xhigh are available, with none as the default so that routine calls stay fast and inexpensive. Choose mini for cost, the full model for harder reasoning.
How much does GPT-5.4 mini cost?
On TokenLab, GPT-5.4 mini costs Input $0.525 / Output $3.15 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 mini?
GPT-5.4 mini accepts up to 400,000 tokens of context and returns up to 128,000 tokens in one response.
Which endpoint should GPT-5.4 mini use?
Use https://api.tokenlab.sh/v1/responses for GPT-5.4 mini. The request example below shows the matching code shape.
Which operations does GPT-5.4 mini support?
GPT-5.4 mini supports Text to text. Select an operation above to see its endpoint and request example.
Compare GPT-5.4 mini
Sources
Reviewed Oct 2, 2026
More from OpenAI