# GPT-5.4 mini

GPT-5.4 mini is the faster, more efficient tier of OpenAI's GPT-5.4 family, designed for high-volume workloads. OpenAI calls it its strongest mini model yet for coding, computer use, and subagents, keeping text and image input and strong reasoning while sitting between GPT-5.4 and the lighter nano. Its published context window is 400,000 tokens. The advertised input rate is $0.75 per million input tokens. The advertised output rate is $4.50 per million output tokens.

- Model identifier: ` openai/gpt-5.4-mini `
- Creator: OpenAI
- Providers: Azure OpenAI, Azure AI Foundry, OpenAI
- Model type: Text generation
- Canonical model page: [GPT-5.4 mini](<https://impossibl.com/openai/gpt-5.4-mini>)
- API availability: Listed as serving through Impossibl.
- Creator release date: 2026-03-17

## Providers

Different companies host GPT-5.4 mini. Impossibl routes your request intelligently and handles failover automatically when a provider is down.

| Provider | Endpoint location | Routes | Verified context window | Tool calling |
| --- | --- | --- | --- | --- |
| Azure OpenAI | East US | 1 | Not verified | Not verified |
| Azure AI Foundry | East US 2 | 1 | Not verified | Not verified |
| OpenAI | Not specified | 1 | Not verified | Not verified |

The model price applies whichever provider serves the request. Endpoint locations do not guarantee inference residency.

## Specs and Capabilities

- Context window: 400,000 tokens
- Maximum output tokens: Not published in the public catalog; this is separate from the context window.
- Inputs: text, image
- Outputs: text
- Tool calling: Not published

## API access

Use ` openai/gpt-5.4-mini ` as the model identifier through the Impossibl API.

Published API endpoints:

- ` /v1/chat/completions `

[API quickstart](<https://impossibl.com/docs/agent-quickstart>) · [Models and billing documentation](<https://impossibl.com/docs/models>)

## Published API prices

All rates are in USD. An unpublished rate is unknown, not zero. Workspace-specific pricing, taxes, balance-purchase fees, billing adjustments, and tool charges can affect the amount paid.

| Usage | Published base rate | Billing unit |
| --- | --- | --- |
| Input | $0.75 | per million input tokens |
| Output | $4.50 | per million output tokens |
| Cache read | $0.075 | per million cached input tokens |
| Cache write (default retention) | Not published | per million tokens |
| Cache write (one-hour retention) | Not published | per million tokens |
| Audio input | Not published | per million audio input tokens |

## Benchmarks

Last checked: 2026-11-12T20:06:14.004Z.
Configurations are evaluated separately. API scores retain their original units. Unmeasured scores are omitted.

### GPT-5.4 mini (Xhigh)

Source: [Evaluation source](<https://artificialanalysis.ai/models/gpt-5-4-mini>)

| Benchmark | Score |
| --- | --- |
| GPQA | 87.47% |
| Humanity’s Last Exam | 28.13% |
| SciCode | 52.08% |
| CritPt | 10.00% |
| Terminal-Bench Hard | 52.27% |
| AA-LCR | 77.00% |
| Multilingual LCR | 20.00% |
| GDPval Elo | 1021.79 |
| GDP.pdf All-pass | 13.40% |
| AutomationBench Partial | 26.20% |
| AnalystAgent | 15.00% |
| MMMU-Pro | 73.29% |
| IFBench | 73.27% |
| Tau2 | 83.33% |
| Tau Banking | 25.57% |
| APEX-Agents | 28.17% |
| ITBench | 35.22% |
| legal index (API) | 26.2 |
| coding index (API) | 56.1 |
| agentic index (API) | 17.9 |
| economics index (API) | 29.7 |
| engineering index (API) | 24.4 |
| intelligence index (API) | 24.1 |
| strategy and ops index (API) | 25.3 |
| finance and accounting index (API) | 24.2 |
| healthcare and medical index (API) | 22.1 |

Full benchmark data and breakdowns: [Complete benchmark snapshot](<https://api.impossibl.com/v1/models/openai/gpt-5.4-mini/benchmarks>).


## Availability and retirement

The model is listed as serving. No retirement is announced in the current public catalog.

## Frequently asked questions

### What is GPT-5.4 mini?

GPT-5.4 mini is the faster, more efficient tier of OpenAI's GPT-5.4 family, designed for high-volume workloads. OpenAI calls it its strongest mini model yet for coding, computer use, and subagents, keeping text and image input and strong reasoning while sitting between GPT-5.4 and the lighter nano.

Sources: [OpenRouter model listing](<https://openrouter.ai/openai/gpt-5.4-mini>); [OpenAI model documentation](<https://developers.openai.com/api/docs/models/gpt-5.4-mini>)

### How much does GPT-5.4 mini cost?

GPT-5.4 mini costs $0.75 per million input tokens and $4.50 per million output tokens through the Impossibl API. Cache reads cost $0.075 per million tokens. All rates are in USD.

### What is the context length of GPT-5.4 mini?

GPT-5.4 mini has a 400,000-token context window through the Impossibl API.

### What inputs and outputs does GPT-5.4 mini support?

GPT-5.4 mini accepts text and image as input. GPT-5.4 mini returns text.

### How do I use GPT-5.4 mini through an API?

GPT-5.4 mini is available through the Impossibl API. Use openai/gpt-5.4-mini as the model identifier with /v1/chat/completions. The quickstart explains how to create an API key and send your first request.

### When was GPT-5.4 mini released?

OpenAI released GPT-5.4 mini on March 17, 2026.

Sources: [Creator release announcement](<https://openai.com/index/introducing-gpt-5-4-mini-and-nano/>)

## Sources and related links

Prices, limits, capabilities, endpoints, and availability come from the public Impossibl catalog. Creator documentation supplies descriptions and release dates where available; it does not override the published API offering.

- [Public Impossibl model catalog](<https://api.impossibl.com/v1/models>)
- [Full model directory as JSON](<https://impossibl.com/models.json>)
- [GPT-5.4 mini: model details and pricing](<https://impossibl.com/openai/gpt-5.4-mini>)
- [Model directory](<https://impossibl.com/models>)
- [OpenRouter model listing](<https://openrouter.ai/openai/gpt-5.4-mini>)
- [OpenAI model documentation](<https://developers.openai.com/api/docs/models/gpt-5.4-mini>)
