# Alibaba Releases Qwen3-VL, an 8B Open-Source Vision Model

> The latest vision-language model from the popular Qwen series is instruction-tuned and available under an Apache 2.0 license.

Published by The Open Weights on Oct 11, 2025. Canonical: https://theopenweights.com/news/qwen3-vl-8b-15l9

## Key facts

- Company: Qwen · Alibaba
- Model: Qwen3-VL-8B-Instruct
- Version: 3-VL
- Category: Vision-Language
- Modalities: Vision-Language
- License: Apache 2.0 (Open weights, commercial use allowed)
- Parameters: 8B
- Context window: 262K tokens
- Architecture: Qwen3VLForConditionalGeneration
- Precision: BF16
- Weights size: 17.5 GB
- Significance: notable
- Published: 2025-10-11
- Last verified: 2026-06-17
- Hugging Face: https://huggingface.co/Qwen/Qwen3-VL-8B-Instruct
- Canonical URL: https://theopenweights.com/news/qwen3-vl-8b-15l9

Alibaba's Qwen team has launched Qwen3-VL-8B-Instruct, a new vision-language model (VLM) built on their latest Qwen3 architecture. This release adds powerful multimodal capabilities to the recently introduced Qwen3 family of open-source models.

As an instruction-tuned VLM, Qwen3-VL-8B is designed to understand and process both images and text simultaneously. It can perform a wide range of tasks that require visual reasoning, such as answering detailed questions about an image, generating captions, and identifying specific objects within a scene.

With 8 billion parameters, the model occupies a practical middle ground, offering strong performance without the demanding hardware requirements of much larger, proprietary systems. Its release provides developers and researchers with a capable tool for building multimodal applications, from enhanced chatbots to sophisticated content analysis systems.

The model is available under the permissive Apache 2.0 license, encouraging both academic and commercial use. Interested users can find the model weights and documentation on the [Qwen Hugging Face repository](https://huggingface.co/Qwen/Qwen3-VL-8B-Instruct).

## Get the model

- [Hugging Face](https://huggingface.co/Qwen/Qwen3-VL-8B-Instruct)

## Sources

- [Qwen/Qwen3-VL-8B-Instruct](https://huggingface.co/Qwen/Qwen3-VL-8B-Instruct) — Hugging Face, Oct 11, 2025

---
Source: The Open Weights (https://theopenweights.com/). Aggregated and written by Claude, curated by humans. Cite as: "Alibaba Releases Qwen3-VL, an 8B Open-Source Vision Model", The Open Weights, Oct 11, 2025, https://theopenweights.com/news/qwen3-vl-8b-15l9