# Alibaba: qwen3.5-plus

- Model ID: `qwen3.5-plus`
- Provider: Alibaba
- Web version: https://www.moleapi.com/en/models/qwen/qwen3.5-plus
- Content status: verified

## Model introduction

Qwen3.5-Plus is listed as an advanced model in the Qwen Code documentation, with thinking enabled.

## Model capabilities

- OpenAI compatible
- Prompt cache
- Reasoning

### Verified specifications

- Official model parameter: qwen3.5-plus
- Verified capability: Thinking enabled
- Official use context: Qwen Code model-provider configuration

## Model pricing and access

Pricing is supplied dynamically by the MoleAPI console API.

### Standard (default, x1)

- Input: $0.12 / 1M tokens
- Output: $0.7 / 1M tokens
- Cache read: $0.012 / 1M tokens
- Cache write: $0.15 / 1M tokens

### Access protocols

| Protocol | Method | Endpoint |
| --- | --- | --- |
| openai | POST | /v1/chat/completions |

- Live pricing source: https://home.moleapi.com/api/pricing

## About qwen3.5-plus

Qwen3.5-Plus is a commercial Qwen3.5 model provided by Alibaba Cloud Model Studio. It supports text, image, and video input and produces text output. The model provides a 1 million token context window, 64K maximum output, function calling, built-in tools, and structured output.[1] It uses a hybrid thinking mode, with thinking enabled by default, and allows switching between thinking and non-thinking responses per request.[2]

## Model Positioning and Core Capabilities

Qwen3.5-Plus combines long context, multimodal understanding, reasoning, and tool calling in a single model. Alibaba Cloud's model table lists it as a commercial model in the Qwen3.5 Plus series and provides the fixed snapshot `qwen3.5-plus-2026-02-15`.[1][2]

The model can call functions based on image or video content and can also use built-in tools such as web search and code execution. In non-thinking mode, it supports structured results that conform to the JSON specification.[1]

## Capabilities and Specifications

Qwen3.5-Plus accepts text, images, and video and returns text. Its context window is 1 million tokens, and the maximum output per request is 64K tokens. A single request can include up to 256 images and 64 videos. Videos are limited to 2 hours or 2 GB.[1]

| Item | Specification |
| --- | --- |
| Input | Text, images, video [1] |
| Output | Text [1] |
| Context window | 1M tokens [1] |
| Maximum output | 64K tokens [1] |
| Images per request | Up to 256 [1] |
| Videos per request | Up to 64 [1] |
| Video limits | Up to 2 hours, up to 2 GB [1] |
| Tool capabilities | Function calling, web search, code execution [1] |
| Structured output | Supported in non-thinking mode [1] |
| Thinking mode | Hybrid thinking, enabled by default [2] |

Hybrid thinking is controlled through the `enable_thinking` parameter: when set to `true`, the model reasons before responding; when set to `false`, the model generates a response directly.[2]

## Use Cases

Qwen3.5-Plus's multimodal input can be used for image analysis and video understanding.[1] Its 1 million token context can accommodate long documents or inputs composed of large amounts of material, while function calling, built-in tools, and structured output cover applications that require external actions or standardized results.[1]

Hybrid thinking allows the same model to switch between reasoning before answering and generating an answer directly.[2]

## Important Limitations

The model outputs text only and does not generate images, video, or audio.[1] Structured output is limited to non-thinking mode. Image and video inputs are both subject to file count limits, while videos are also subject to duration and file size limits.[1]

Alibaba Cloud currently places Qwen3.5 in the legacy model group while continuing to list the full specifications for `qwen3.5-plus` and its fixed snapshot.[1]

## Sources

[1] [Alibaba Cloud Model Studio: Visual understanding](https://www.alibabacloud.com/help/en/model-studio/vision-model/)
[2] [Alibaba Cloud Model Studio: Deep thinking](https://www.alibabacloud.com/help/en/model-studio/deep-thinking)



## Code examples

### cURL

```bash
curl https://api.moleapi.com/v1/chat/completions \
  -H "Authorization: Bearer $MOLEAPI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"qwen3.5-plus","messages":[{"role":"user","content":"Hello"}]}'
```

## More related models

- [qwen3-next-80b-a3b-instruct](https://www.moleapi.com/en/models/qwen/qwen3-next-80b-a3b-instruct) — Alibaba
- [deepseek-r1-0528-qwen3-8b](https://www.moleapi.com/en/models/qwen/deepseek-r1-0528-qwen3-8b) — Alibaba
- [deepseek-r1-distill-qwen-1.5b](https://www.moleapi.com/en/models/qwen/deepseek-r1-distill-qwen-1.5b) — Alibaba

## Frequently asked questions

### How is qwen3.5-plus priced?

Prices are read from the MoleAPI console API and update with model prices, context tiers, and account groups.

### How can I access qwen3.5-plus?

The catalog currently lists openai.

### How do I switch an existing project to qwen3.5-plus?

Keep the MoleAPI API address and key, replace the model parameter, and check protocol-specific parameters.

### Where do the qwen3.5-plus details come from?

Capabilities and limitations are checked against Qwen and the other cited pages; pricing and protocols come from the MoleAPI console.