# Qwen2.5 VL AWQ 3B **Qwen2.5 VL AWQ 3B** is a vision-language model with 3 billion parameters. It reads text and images, returns text, and handles visual analysis, agentic reasoning, long video comprehension, visual localization, and structured output generation. [AI Inference](/en/documentation/platform/ai-inference/) runs it under the id `qwen-qwen25-vl-3b-instruct-awq`. ## Model details The id is what a request sends to reach this model, and `Azion.AI.run` takes it as the first argument. The HuggingFace repository holds the model card. | Detail | Value | | -------------------------- | ----------------------------------------------------------------------------------------- | | Model name | Qwen2.5 VL | | Version | AWQ 3B | | Model category | Vision-Language Model (VLM) | | Model id | `qwen-qwen25-vl-3b-instruct-awq` | | Size | 3B parameters | | HuggingFace model | [Qwen/Qwen2.5-VL-3B-Instruct-AWQ](https://huggingface.co/Qwen/Qwen2.5-VL-3B-Instruct-AWQ) | | OpenAI-compatible endpoint | [OpenAI Chat API](https://developers.openai.com/api/reference/overview) | | License | [Apache 2.0](https://choosealicense.com/licenses/apache-2.0/) | ## Capabilities These capabilities are properties of the model, and a request body does not set them. For every field a chat request accepts, with its type, default, and bounds, refer to [Model invocation](/en/documentation/platform/ai-inference/model-invocation/). | Capability | Value | | -------------- | -------------- | | Input data | Text and image | | Context length | 32k tokens | | Tool calling | Yes | | Supports LoRA | Yes | --- ## Usage You call this model from a [function](/en/documentation/platform/functions/), through the `Azion.AI.run` binding. The first argument is the model id, and the second is an OpenAI-compatible request body. Each example on this page uses that binding. For the HTTP form of the same request, refer to [Model invocation](/en/documentation/platform/ai-inference/model-invocation/). ### Chat completion This request carries a system message that sets the behavior of the model and a user message that holds the prompt: ```ts const modelResponse = await Azion.AI.run("qwen-qwen25-vl-3b-instruct-awq", { "stream": false, "messages": [ { "role": "system", "content": "You are a helpful assistant." }, { "role": "user", "content": "Name the European capitals." } ] }) ``` One entry arrives in `choices`, and the text the model generated sits at `choices[0].message.content`: ```json { "id": "chatcmpl-0123456789abcdef0123456789abcdef", "object": "chat.completion", "created": 1767268800, "model": "qwen-qwen25-vl-3b-instruct-awq", "choices": [ { "index": 0, "message": { "role": "assistant", "reasoning_content": null, "content": "Sure! Here is a list of some European capitals...", "tool_calls": [] }, "logprobs": null, "finish_reason": "stop", "stop_reason": null } ], "usage": { "prompt_tokens": 9, "total_tokens": 527, "completion_tokens": 518, "prompt_tokens_details": null }, "prompt_logprobs": null } ``` ### Tool calling A `tools` array declares the functions the model may call. Every entry sets `type` to `function` and carries a `function` object with three fields: the `name` the model returns, a `description` of what the function does, and the `parameters` it accepts, written as a JSON Schema object: ```ts const modelResponse = await Azion.AI.run("qwen-qwen25-vl-3b-instruct-awq", { "stream": false, "messages": [ { "role": "system", "content": "You are a helpful assistant with access to tools." }, { "role": "user", "content": "What is the weather in London?" } ], "tools": [ { "type": "function", "function": { "name": "get_weather", "description": "Get the current weather for a location", "parameters": { "type": "object", "properties": { "location": { "type": "string", "description": "The city and state" } }, "required": ["location"] } } } ] }) ``` A response that selects a tool carries `null` in `content`, one entry in `tool_calls` holding the function name and the arguments the model chose, and `tool_calls` in `finish_reason`: ```json { "id": "chatcmpl-fedcba9876543210fedcba9876543210", "object": "chat.completion", "created": 1746821866, "model": "qwen-qwen25-vl-3b-instruct-awq", "choices": [ { "index": 0, "message": { "role": "assistant", "reasoning_content": null, "content": null, "tool_calls": [ { "id": "chatcmpl-tool-0123456789abcdef0123456789abcdef", "type": "function", "function": { "name": "get_weather", "arguments": "{\"location\": \"London\"}" } } ] }, "logprobs": null, "finish_reason": "tool_calls", "stop_reason": null } ], "usage": { "prompt_tokens": 293, "total_tokens": 313, "completion_tokens": 20, "prompt_tokens_details": null }, "prompt_logprobs": null } ``` ### Multimodal input A user message that carries an image sets `content` to an array of parts instead of a string. Each part states its `type`: a `text` part holds the prompt, and an `image_url` part holds the URL the model reads the image from. For every content part a message accepts, refer to [Message objects](/en/documentation/platform/ai-inference/model-invocation/#message-objects). This request sends one text part and one image part inside the same user message: ```ts const modelResponse = await Azion.AI.run("qwen-qwen25-vl-3b-instruct-awq", { "stream": false, "messages": [ { "role": "system", "content": "You are a helpful assistant." }, { "role": "user", "content": [ { "type": "text", "text": "What is in this image?" }, { "type": "image_url", "image_url": { "url": "https://example.com/image.jpg" } } ] } ] }) ``` The model returns the same response object as a chat completion, and what it read in the image sits at `choices[0].message.content`. --- ## Related resources - [Model invocation](/en/documentation/platform/ai-inference/model-invocation.md): The request fields, the message objects, and the HTTP endpoint that takes the same body. - [AI models](/en/documentation/platform/ai-inference/models.md): The catalog of models AI Inference runs, with the id and the capabilities of each one. - [AI Inference](/en/documentation/platform/ai-inference.md): The product that loads and runs this model. - [Functions](/en/documentation/platform/functions.md): The product whose runtime carries the binding these examples call. - [LoRA Fine-Tune](/en/documentation/platform/ai-inference/lora-fine-tune.md): The extension of AI Inference that adapts a model with LoRA, and the models that accept it. - [Glossary](/en/documentation/platform/ai-inference/glossary.md): Where model id, context length, and the rest of the AI Inference vocabulary are defined.