# QVQ-72B-Preview

> Alibaba's experimental open-weight visual-reasoning model: built on Qwen2-VL-72B, it 'thinks out loud' step by step over images. Research preview, December 2024.

- Maker: Alibaba (Qwen)
- Line: QwQ / QVQ (reasoning preview)
- Canonical page: https://ai-tldr.dev/models/qvq-72b-preview/
- Updated: 2026-06-18

## In short

QVQ-72B-Preview is an experimental open-weight visual-reasoning model released by Alibaba's Qwen team on 25 December 2024. You can use it through Hugging Face (download weights / Transformers).

## Facts

| Fact | Value | Source |
| --- | --- | --- |
| Context window | ~32K (inherited from the Qwen2-VL-72B base; QVQ's own card does not state a separate figure) (max output Not separately specified — the model card's example uses max_new_tokens=8192; long reasoning traces consume the output budget) | [1] |
| Licence | Qwen License (license: other, license_name: qwen) — open weights, commercial use permitted · open weights | [1] |
| Released | 25 December 2024 | [2] |
| Parameters | ~73B (built on the Qwen2-VL-72B base; the name uses the 72B base size, aggregators report ~73B with the vision encoder) | [1] |
| Knowledge cutoff | Not officially disclosed (inherits the Qwen2-VL-72B base) | [1] |

## Sources

1. [Qwen/QVQ-72B-Preview — Hugging Face model card (benchmarks, license, limitations)](https://huggingface.co/Qwen/QVQ-72B-Preview)
2. [QVQ: To See the World with Wisdom — official Qwen blog](https://qwenlm.github.io/blog/qvq-72b-preview/)

## Overview

QVQ-72B-Preview is an experimental open-weight visual-reasoning model released by Alibaba's Qwen team on 25 December 2024. It is built on the Qwen2-VL-72B vision-language model and adds a chain-of-thought 'thinking' stage: given an image and a question, QVQ streams a long, step-by-step reasoning trace before settling on an answer — the visual counterpart to a text reasoning model like OpenAI o1. Qwen positioned it as a research preview to explore reasoning over non-textual (visual) inputs, and shipped the weights publicly on Hugging Face and GitHub.

On Qwen's reported benchmarks, QVQ-72B-Preview scores 70.3% on MMMU (val), a clear lift over its Qwen2-VL-72B base, with strong math-and-vision results on MathVista (71.4%), MathVision (35.9%), and OlympiadBench (20.4%). Qwen framed it as narrowing the gap with the leading o1 model on multimodal reasoning. It was first published with an Apache-2.0 tag that the team quickly corrected to the Qwen License; the weights remain openly downloadable with commercial use permitted.

QVQ-72B-Preview was explicitly experimental. The model card warns of language mixing and code-switching, recursive reasoning loops that produce verbose output, and a tendency to lose focus on the image during long reasoning and hallucinate. It supports only single-round dialogue and does not accept video, and Qwen said it could not fully replace Qwen2-VL-72B-Instruct. It sits in Qwen's QwQ/QVQ reasoning-preview line alongside QwQ-32B-Preview; that work later matured into QwQ-32B (March 2025) and was ultimately merged into the unified Qwen3 'thinking' family, so QVQ-72B-Preview is best understood as a historical milestone rather than a model to deploy today.

## Benchmarks

- MMMU (val): 70.3% ([source](https://huggingface.co/Qwen/QVQ-72B-Preview))
- MathVista (mini): 71.4% ([source](https://huggingface.co/Qwen/QVQ-72B-Preview))
- MathVision (full): 35.9% ([source](https://huggingface.co/Qwen/QVQ-72B-Preview))
- OlympiadBench: 20.4% ([source](https://huggingface.co/Qwen/QVQ-72B-Preview))

## Strengths

- First widely available open-weight VISUAL reasoning model — 'thinks out loud' step by step over an image, not just over text
- Strong multimodal-reasoning benchmarks for late 2024: MMMU 70.3% (near Claude 3.5 Sonnet) and MathVista 71.4% (ahead of the reported GPT-4o and Claude 3.5 Sonnet figures)
- Clear gains over its own Qwen2-VL-72B base on math-and-vision tasks (e.g. MathVision 35.9% vs 25.9%, OlympiadBench 20.4% vs 11.2%)
- Open weights under the Qwen License with commercial use permitted — downloadable on Hugging Face and GitHub for self-hosting and research
- Transparent, inspectable chain-of-thought makes its visual reasoning easy to study and debug

## Best for

- Research into multimodal / visual chain-of-thought reasoning
- Step-by-step reasoning over diagrams, charts, math problems, and figures in images
- Visual math and science problem-solving (MathVista / MathVision / OlympiadBench-style tasks)
- Self-hosted experimentation with an open-weight visual reasoning model
- Baseline or teacher for studying and distilling visual reasoning traces

## How to access

- [Hugging Face (download weights / Transformers)](https://huggingface.co/Qwen/QVQ-72B-Preview) — model id `Qwen/QVQ-72B-Preview`

## QwQ / QVQ (reasoning preview) versions

- [QwQ-32B](https://ai-tldr.dev/models/qwq-32b/) (5 March 2025): Production 32B text reasoning model; the line's current release
- QVQ-72B-Preview (25 December 2024): Experimental visual-reasoning preview built on Qwen2-VL-72B — this page

## FAQ

### What is QVQ-72B-Preview?

QVQ-72B-Preview is an experimental open-weight visual-reasoning model released by Alibaba's Qwen team on 25 December 2024. Built on Qwen2-VL-72B, it adds a chain-of-thought 'thinking' stage: given an image and a question, it streams a long step-by-step reasoning trace before answering — effectively a visual counterpart to text reasoning models like OpenAI o1. It was published as a research preview, not a production model.

### How well does QVQ-72B-Preview perform on benchmarks?

On the numbers Qwen reported on the model card, QVQ-72B-Preview scores 70.3% on MMMU (val), 71.4% on MathVista (mini), 35.9% on MathVision (full), and 20.4% on OlympiadBench. These are clear gains over its Qwen2-VL-72B base on visual-math tasks, and Qwen described the model as narrowing the gap with the leading o1 model on multimodal reasoning.

### Is QVQ-72B-Preview open source, and what is its license?

The weights are open and downloadable on Hugging Face and GitHub. The license is the Qwen License (tagged license: other, license_name: qwen) and permits commercial use. It was briefly published with an Apache-2.0 tag at launch, which the Qwen team quickly corrected to the Qwen License.

### Should I use QVQ-72B-Preview today, or something newer?

QVQ-72B-Preview is a historical research preview with known limitations — language mixing, recursive reasoning loops, single-round-only dialogue, no video input, and a tendency to lose track of the image and hallucinate during long reasoning. Qwen's reasoning work has since moved on through QwQ-32B (March 2025) and into the unified Qwen3 'thinking' family, so for production visual reasoning a current Qwen (or other up-to-date) model is the better choice.

## Links

- [QVQ: To See the World with Wisdom — official Qwen blog](https://qwenlm.github.io/blog/qvq-72b-preview/)
- [Qwen/QVQ-72B-Preview — Hugging Face model card (benchmarks, license, limitations)](https://huggingface.co/Qwen/QVQ-72B-Preview)
- [Qwen — official site](https://qwen.ai)
- [Simon Willison — Trying out QvQ, Qwen's new visual reasoning model](https://simonwillison.net/2024/Dec/24/qvq/)
- [SiliconANGLE — Alibaba announces experimental visual reasoning QVQ-72B model](https://siliconangle.com/2024/12/26/alibaba-announces-advanced-experimental-visual-reasoning-qvq-72b-ai-model/)

Source: AI/TLDR LLM registry, https://ai-tldr.dev/models/qvq-72b-preview/