Visual document searchSearch and RAG

ColPali / ColQwen

Searches PDFs and scans as images: pages do not need to be OCR'd first, the model finds the right one for a question directly, including tables and charts. Trained on English.

Developer: Illuin Technology (ViDoRe team), France
First release: Jun 2024
Latest release: Jul 2025
Sizes: 256M – 3B
License: Commercial use with conditionsAdapters are MIT or Apache 2.0, but the base models have their own licenses: PaliGemma — Gemma, Qwen2.5-VL-3B — Qwen research license (ColQwen2 on Qwen2-VL — Apache 2.0)
Russian: Not supported
Running: On your own serverAlso runs without a GPU
Industries: Documents and accounting, Legal, Finance, Software development

Deploy on your server Hugging Face GitHub

What it does

Search across scans, presentations and PDFs
RAG over documents with tables and charts
Search across technical documentation

Where it is used

Document managementLaw firmsBanking and insurance

Hardware requirements

LaptopLaptop or regular PC, up to 8 GB of VRAM — smaller versions

fits

1 GPUOne GPU with 16–80 GB — mid-size versions

fits

ClusterServer with several GPUs — flagship versions

no versions

Versions

Jul 2025ColQwen-Omni
Jan 2025ColSmol 256M и 500M, ColQwen2.5
Nov 2024ColPali v1.3 и ColQwen2 v1.0
Sep 2024ColQwen2
Aug 2024ColPali v1.2
Jun 2024ColPali

How to run it

On your own serverWeights are downloaded from Hugging Face and served with vLLM (load and API) or llama.cpp (modest hardware). The model then runs in a closed network with no per-request fees.Hugging Face

How much hardware you needCalculate the VRAM for the model size, context length and number of concurrent requests.Open the hardware calculator

I can set this up end to end: pick the model size, deploy it on your server and connect it to your systems.

Frequently asked questions

Can ColPali / ColQwen be used in a commercial project?

With conditions. License: Adapters are MIT or Apache 2.0, but the base models have their own licenses: PaliGemma — Gemma, Qwen2.5-VL-3B — Qwen research license (ColQwen2 on Qwen2-VL — Apache 2.0). Restrictions vary — region, company revenue, attribution requirements. Have a lawyer check the terms before a commercial launch.

What hardware does ColPali / ColQwen need?

At minimum: Laptop or regular PC, up to 8 GB of VRAM — smaller versions. Some versions also run on an ordinary CPU, without a GPU. You can calculate the exact VRAM for your model size and context in the hardware calculator.

Does ColPali / ColQwen support Russian?

No. The model card lists its languages and Russian is not among them.

Where can I download ColPali / ColQwen and what does it cost?

The ColPali / ColQwen weights are open and free to download. You only pay for the hardware it runs on and for the setup. Source links are at the bottom of this page.

How I deploy it for clients

SelectionI pick the model size for your task and hardware and test it on your examples.
DeploymentI deploy it on your server or in a closed network and provide an API.
Fine-tuningI fine-tune it on your data (LoRA) or connect a knowledge base — whichever is cheaper for the task.
IntegrationI connect it to your CRM, ERP, bot, website or team chat and set up monitoring.

Embeddings and rerankers based on Qwen3, among the best open ones for multilingual search, including Russian. VL versions search images, screenshots and video.

Details Search and RAGJina EmbeddingsJina AI · GermanyCommercial use with conditions

Strong multilingual embeddings with long context; v5-omni understands text, images and audio. Recent versions are open for non-commercial use only.

Details Search and RAGNomic EmbedNomic AI · USACommercial use allowed

Fully open embeddings, with weights, data and training code. v2 is multilingual on MoE; there are versions for code and for searching PDF pages.

Details

Source: huggingface.co/vidore/colqwen-omni-v0.1. Data checked against the model card on 22 Sep 2026. Have a lawyer review the license before commercial launch.