# Qwen3.6-35B-A3B

> Open-weight multimodal model from Alibaba Cloud that activates a small share of its parameters per token.

## Key facts

- Model author: Qwen
- Input: $0.375 per 1M tokens
- Output: $2.25 per 1M tokens
- Context window: 256,000 tokens
- Maximum output: 64,000 tokens
- Input modalities: image, text, video
- Output modalities: text
- Capabilities: structured_output, tools, vision
- Released: 2026-04-27
- Model IDs: `qwen/qwen3.6-35b-a3b`
- Model page: https://www.realrelay.ai/model/qwen/qwen3.6-35b-a3b

## Overview

Qwen3.6-35B-A3B is an open-weight multimodal model from Alibaba Cloud with 35 billion total parameters and 3 billion active parameters per token. It uses a hybrid sparse mixture-of-experts architecture, accepts image, text and video input, and supports tool calling and structured output.

It has a 256K-token context window and can generate up to 64,000 output tokens. It accepts text, image and video input and returns text. It was released on April 27, 2026. Its weights are open.

## Capabilities

Qwen3.6-35B-A3B supports tool calling, structured output and vision. It accepts text, image and video input and returns text.

Only a small share of the parameters is active for each token, which is the point of the sparse design. The model supports tool calling, structured output and vision, and Alibaba documents function calling that acts on image or video content.

## Pricing

Qwen3.6-35B-A3B costs $0.375 per 1M input tokens and $2.25 per 1M output tokens. Qwen3.6-35B-A3B is available through RealRelay's OpenAI-compatible API.

## Use cases

Use Qwen3.6-35B-A3B for multimodal agents that read screenshots, photos or video and call tools in response, with a 256K-token context window for long sessions.

Its open weights and small active parameter count suit teams that want a vision-capable model they can prototype against the hosted API and later self-host, returning structured output for downstream code.

## 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/qwen3.6-35b-a3b", "messages": [{"role": "user", "content": "Hello"}]}'
```

## FAQ

### What is Qwen3.6-35B-A3B?

Qwen3.6-35B-A3B is an open-weight multimodal model from Alibaba Cloud with 35 billion total parameters and 3 billion active parameters per token.

### How much does Qwen3.6-35B-A3B cost?

Qwen3.6-35B-A3B costs $0.375 per 1M input tokens and $2.25 per 1M output tokens.

### What is the context window of Qwen3.6-35B-A3B?

Qwen3.6-35B-A3B has a 256K-token context window and can generate up to 64,000 output tokens.

### What inputs and outputs does Qwen3.6-35B-A3B support?

Qwen3.6-35B-A3B accepts text, image and video input and returns text.

### Does Qwen3.6-35B-A3B support tool calling?

Yes. Qwen3.6-35B-A3B supports tool calling, structured output and vision.

### When was Qwen3.6-35B-A3B released?

Qwen3.6-35B-A3B was released on April 27, 2026.

### How do I call Qwen3.6-35B-A3B via the API?

Send requests to RealRelay's OpenAI-compatible API with the model ID qwen/qwen3.6-35b-a3b to use Qwen3.6-35B-A3B.