# GPT-4.1

> OpenAI's non-reasoning model for instruction following, coding and long documents, with image input.

## Key facts

- Model author: OpenAI
- Input: $2.00 per 1M tokens
- Output: $8.00 per 1M tokens
- Cached input: $0.50 per 1M tokens
- Context window: 1,047,576 tokens
- Maximum output: 32,768 tokens
- Input modalities: image, text
- Output modalities: text
- Capabilities: fine_tuning, structured_output, tools, vision
- Released: 2025-04-14
- Knowledge cutoff: Jun 01, 2024
- Model IDs: `openai/gpt-4.1`
- Model page: https://www.realrelay.ai/model/openai/gpt-4.1

## Overview

GPT-4.1 is a non-reasoning model from OpenAI, built for instruction following, coding and long-context work. It accepts text and image input and returns text, supports tool calling and structured output, and offers a context window of 1,047,576 tokens. It answers directly rather than through a reasoning phase, which suits applications that need predictable output.

GPT-4.1 has a context window of 1,047,576 tokens and can generate up to 32,768 output tokens. It accepts text and image input and returns text output. Its knowledge cutoff is June 1, 2024, and it was released on April 14, 2025.

## Capabilities

GPT-4.1 supports tool calling, structured output and image input, and it returns text. It can generate up to 32,768 output tokens per request.

As a non-reasoning model, it answers directly instead of spending output on a reasoning phase. Cached input reads are priced separately.

## Pricing

GPT-4.1 is billed per token: $2.00 per 1M input tokens and $8.00 per 1M output tokens. Cached input reads cost $0.50 per 1M tokens. GPT-4.1 is available through RealRelay's OpenAI-compatible API.

## Use cases

Use GPT-4.1 for coding assistants that read large files or whole repositories, follow detailed instructions and call tools, using the 1,047,576-token window to keep the code in one request.

It also suits long-document work, such as answering questions over contracts, reports or screenshots, and returning structured output that other systems can parse.

## Call it

OpenAI-compatible chat completions:

```bash
curl https://www.realrelay.ai/v1/chat/completions \
  -H "Authorization: Bearer $REALRELAY_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "openai/gpt-4.1", "messages": [{"role": "user", "content": "Hello"}]}'
```

## FAQ

### What is GPT-4.1?

GPT-4.1 is a non-reasoning model from OpenAI, built for instruction following, coding and long-context work. It accepts text and image input and returns text, supports tool calling and structured output, and offers a context window of 1,047,576 tokens. It answers directly rather than through a reasoning phase, which suits applications that need predictable output.

### How much does GPT-4.1 cost?

GPT-4.1 costs $2.00 per 1M input tokens and $8.00 per 1M output tokens, with cached input reads at $0.50 per 1M tokens.

### What is the context window of GPT-4.1?

GPT-4.1 has a context window of 1,047,576 tokens and can generate up to 32,768 output tokens.

### What inputs and outputs does GPT-4.1 support?

GPT-4.1 accepts text and image input and returns text output.

### Which features does GPT-4.1 support?

GPT-4.1 supports tool calling, structured output and image input.

### When was GPT-4.1 released?

GPT-4.1 was released on April 14, 2025. Its knowledge cutoff is June 1, 2024.

### How do I call GPT-4.1 through the API?

To call GPT-4.1, send a request to RealRelay's OpenAI-compatible API with the model ID openai/gpt-4.1.