# Qwen3.5-122B-A10B

> Open-weight native vision-language model from Alibaba that mixes linear attention with sparse experts.

## Key facts

- Model author: Qwen
- Input: $0.40 per 1M tokens
- Output: $3.20 per 1M tokens
- Context window: 256,000 tokens
- Maximum output: 8,000 tokens
- Input modalities: image, text, video
- Output modalities: text
- Capabilities: reasoning, structured_output, tools, vision
- Released: 2026-02-25
- Model IDs: `qwen/qwen3.5-122b-a10b`
- Model page: https://www.realrelay.ai/model/qwen/qwen3.5-122b-a10b

## Overview

Qwen3.5-122B-A10B is an open-weight native vision-language model from Alibaba. It is built on a hybrid architecture that combines a linear attention mechanism with a sparse mixture-of-experts model, which Alibaba says gives it higher inference efficiency, and it accepts image, text and video input.

It has a 256K-token context window and can generate up to 8,000 output tokens. It accepts text, image and video input and returns text. It was released on February 25, 2026. Its weights are open.

## Capabilities

Qwen3.5-122B-A10B supports tool calling, reasoning, structured output and vision. It accepts text, image and video input and returns text.

As a native vision-language model, it reads images and video with the same model that reads text. Function calling lets it act on image or video content, and in non-thinking mode structured output returns valid JSON from visual inputs.

## Pricing

Qwen3.5-122B-A10B costs $0.40 per 1M input tokens and $3.20 per 1M output tokens. Qwen3.5-122B-A10B is available through RealRelay's OpenAI-compatible API.

## Use cases

Use Qwen3.5-122B-A10B for visual question answering and for reading screenshots, photos and video clips together with a text prompt, with a 256K-token context window for long conversations.

Its open weights, sparse design and structured output suit teams building extraction or agent pipelines, for example pulling product details from photos into JSON, with the option of running the same weights themselves.

## 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/qwen3.5-122b-a10b", "messages": [{"role": "user", "content": "Hello"}]}'
```

## FAQ

### What is Qwen3.5-122B-A10B?

Qwen3.5-122B-A10B is an open-weight native vision-language model from Alibaba.

### How much does Qwen3.5-122B-A10B cost?

Qwen3.5-122B-A10B costs $0.40 per 1M input tokens and $3.20 per 1M output tokens.

### What is the context window of Qwen3.5-122B-A10B?

Qwen3.5-122B-A10B has a 256K-token context window and can generate up to 8,000 output tokens.

### What inputs and outputs does Qwen3.5-122B-A10B support?

Qwen3.5-122B-A10B accepts text, image and video input and returns text.

### Does Qwen3.5-122B-A10B support tool calling?

Yes. Qwen3.5-122B-A10B supports tool calling, reasoning, structured output and vision.

### When was Qwen3.5-122B-A10B released?

Qwen3.5-122B-A10B was released on February 25, 2026.

### How do I call Qwen3.5-122B-A10B via the API?

Send requests to RealRelay's OpenAI-compatible API with the model ID qwen/qwen3.5-122b-a10b to use Qwen3.5-122B-A10B.