# Qwen3-Max

> Alibaba text model focused on reasoning, instruction following, multilingual support and long-tail knowledge.

## Key facts

- Model author: Qwen
- Input: $1.20 per 1M tokens (starting tier)
- Output: $6.00 per 1M tokens (starting tier)
- Cached input: $0.24 per 1M tokens
- Context window: 256,000 tokens
- Input modalities: text
- Output modalities: text
- Capabilities: reasoning, structured_output, tools
- Released: 2025-09-23
- Knowledge cutoff: 2025-06-30
- Model IDs: `qwen/qwen3-max`
- Model page: https://www.realrelay.ai/model/qwen/qwen3-max

## Overview

Qwen3-Max is a language model from Alibaba in the Qwen3 series, focused on reasoning, instruction following, multilingual support and long-tail knowledge coverage. It takes text in and returns text, and supports tool calling and structured output for use inside applications.

It has a 256K-token context window. It accepts text input and returns text. Its knowledge cutoff is June 30, 2025 and it was released on September 23, 2025.

## Capabilities

Qwen3-Max supports tool calling, reasoning and structured output. It accepts text input and returns text.

Beyond reasoning, Alibaba documents the model for instruction following, multilingual support and coverage of long-tail knowledge, meaning less common facts and topics.

## Pricing

Pricing for Qwen3-Max starts from $1.20 per 1M input tokens and $6.00 per 1M output tokens, with cache reads from $0.24 per 1M tokens. Pricing is tiered, so these figures are the starting tier. Qwen3-Max is available through RealRelay's OpenAI-compatible API.

## Use cases

Use Qwen3-Max for multilingual assistants and support chat that must follow detailed instructions and answer questions on less common topics in several languages.

It also fits reasoning-heavy text tasks, such as analysing long documents or planning multi-step work, using tool calling to fetch data and structured output to return results in a fixed format.

## 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": "qwen/qwen3-max", "messages": [{"role": "user", "content": "Hello"}]}'
```

## FAQ

### What is Qwen3-Max?

Qwen3-Max is a language model from Alibaba in the Qwen3 series, focused on reasoning, instruction following, multilingual support and long-tail knowledge coverage.

### How much does Qwen3-Max cost?

Pricing for Qwen3-Max starts from $1.20 per 1M input tokens and $6.00 per 1M output tokens, with cache reads from $0.24 per 1M tokens. Pricing is tiered, so these figures are the starting tier.

### What is the context window of Qwen3-Max?

Qwen3-Max has a 256K-token context window.

### What inputs and outputs does Qwen3-Max support?

Qwen3-Max accepts text input and returns text.

### Does Qwen3-Max support tool calling?

Yes. Qwen3-Max supports tool calling, reasoning and structured output.

### When was Qwen3-Max released?

Qwen3-Max was released on September 23, 2025.

### How do I call Qwen3-Max via the API?

Send requests to RealRelay's OpenAI-compatible API with the model ID qwen/qwen3-max to use Qwen3-Max.