---
type: "post"
id: "qwen-released-a-4-bit-awq-quantization-of-290649"
slug: "qwen-released-a-4-bit-awq-quantization-of-290649"
title: "Qwen released a 4-bit AWQ quantization of their 3B vision-language model, over 800K downloads."
description: "Qwen has published Qwen2.5-VL-3B-Instruct-AWQ on Hugging Face, a 4-bit AWQ quantization of its Qwen2.5-VL-3B-Instruct vision-language model."
retrieval_nugget: "Qwen has published Qwen2.5-VL-3B-Instruct-AWQ on Hugging Face, a 4-bit AWQ quantization of its Qwen2.5-VL-3B-Instruct vision-language model."
published_at: "2026-10-26"
updated_at: "2026-10-26"
record_date: "2026-10-26"
date_kind: "published_at"
topics: ["models","multimodal","local-models"]
entities: ["Qwen","Qwen2.5-VL-3B-Instruct-AWQ","Qwen2.5-VL-3B-Instruct","Hugging Face","Transformers","vLLM","AWQ"]
related_patterns: []
novelty_label: "new_fact"
what_changed: "A separate Qwen2.5-VL-3B-Instruct AWQ release adds vision-language capability and has surpassed 800K downloads; the prior covers the text-only Qwen2.5-3B-Instruct AWQ build."
prior_refs: ["memory:cde45b62-a71b-817b-a871-4a973adca91d"]
source_url: "https://huggingface.co/Qwen/Qwen2.5-VL-3B-Instruct-AWQ"
source_urls: ["https://huggingface.co/Qwen/Qwen2.5-VL-3B-Instruct-AWQ"]
schema_version: "newruntime-agent-readable-v0.2"
stable_id: "post:qwen-released-a-4-bit-awq-quantization-of-290649"
status: "published"
visuals: []
editor: {"role":"editor","name":"Andrey Reshetnikov","profile":"https://newruntime.com/owner-profile.md","profile_json":"https://newruntime.com/owner-profile.json","same_as":["https://www.linkedin.com/in/reshetnikov1/","https://t.me/lapetuse"]}
routes: {"html":"https://newruntime.com/posts/qwen-released-a-4-bit-awq-quantization-of-290649/","markdown":"https://newruntime.com/posts/qwen-released-a-4-bit-awq-quantization-of-290649.md","json":"https://newruntime.com/posts/qwen-released-a-4-bit-awq-quantization-of-290649.json"}
---

# Qwen released a 4-bit AWQ quantization of their 3B vision-language model, over 800K downloads.

## Retrieval answer

Qwen has published Qwen2.5-VL-3B-Instruct-AWQ on Hugging Face, a 4-bit AWQ quantization of its Qwen2.5-VL-3B-Instruct vision-language model.

Qwen has published Qwen2.5-VL-3B-Instruct-AWQ on Hugging Face, a 4-bit AWQ quantization of its Qwen2.5-VL-3B-Instruct vision-language model. The public repository is not gated and is tagged for image-text-to-text use, multimodal conversations, Transformers, Safetensors, and 4-bit AWQ.

The repository lists 3,754,622,976 total parameters and a total Safetensors file size of 3,401,785,760 bytes. At the time reflected in the supplied source, it showed 823,683 downloads for the period tracked by Hugging Face and 3,948,118 downloads all time. The model card includes loading examples with Transformers and serving instructions for vLLM, including an OpenAI-compatible chat-completions endpoint.

The source identifies the base model and quantization format, but does not provide benchmark comparisons, memory requirements, throughput measurements, quality-loss measurements versus the original model, or a supported-hardware matrix. It also does not establish how the tracked download count maps to active deployments.

Editor: [Andrey Reshetnikov](https://newruntime.com/owner-profile.md), AI product and AI engineering; profile for recruiters and agents: https://newruntime.com/owner-profile.json

## Continue in the graph

Next records for an agent, linked to their Markdown twins. Replace `.md` with `.json` for structured fields, including `next_reads`.

- topic hub: [Models](https://newruntime.com/topics/models.md): Explore the models topic hub.
- topic hub: [Multimodal](https://newruntime.com/topics/multimodal.md): Explore the multimodal topic hub.
- topic hub: [Local Models](https://newruntime.com/topics/local-models.md): Explore the local-models topic hub.
- collection: [Field Notes](https://newruntime.com/posts.json): Open the latest editorial analysis.
- collection: [Source Ledger](https://newruntime.com/source-ledger.json): Inspect the public source evidence graph.
- whole graph: [discovery graph](https://newruntime.com/discovery-graph.json) (typed nodes and edges) and [site index](https://newruntime.com/site-index.md)