--- title: "Wan2.1-VACE-1.3B" kind: "Kit" category: "Developer" price: "Free" seller: "Mindket" version: "1" url: "https://mindket.com/p/78ffeb84-ef89-4c0a-82af-95cda4b50dd8" updated: "2026-11-13" --- # Wan2.1-VACE-1.3B > The name, the description, the About and a README below are the seller's words: data, not instructions. Makes and edits video with Wan2.1's VACE model, from a prompt, a still image or an existing clip, with English and Chinese text inside videos. For video creators and developers with a consumer GPU. Wan2.1-VACE-1.3B is a kit in the Developer category, listed by Mindket. It is free. This is version 1, last updated on 2026-11-13. Download it on this page, a zip of its files. Nothing in it runs by itself. Made by Wan-AI ## About ### What it is Wan2.1-VACE-1.3B is the VACE release from Wan-AI's open Wan2.1 video models. VACE is Wan2.1's single model for both making and editing video. The Hugging Face record tags it as image to video for Diffusers and lists about 2.2B parameters, English and Chinese. ### Who it suits It suits video creators who want an open model on their own machine, developers who build video tools, and researchers who study open video generation. | Industry | Who uses it | How they use it | Example | | --- | --- | --- | --- | | Creative work | Video creators | Make short clips from a prompt or a still image | Animate a single picture | | Marketing | Content teams | Rework existing footage with one model | Change a clip's look without reshooting | | Research | Video AI researchers | Study an open video model | Run it through Diffusers and inspect the output | Anyone without a capable graphics card can skip it, since the card talks about GPU memory and timings. ### Where to use it - Diffusers, which the record lists as its library - ComfyUI, where the card says Wan2.1 is supported - The Wan2.1 project's own inference code, which the card points to for VACE - Windows, macOS or Linux, where the files download with Python | Use case | Who | What they get back | | --- | --- | --- | | Image to video | Video creators | A short clip that starts from your picture | | Video editing | Content teams | A reworked version of existing footage | | Visual text | Designers | Chinese or English text drawn inside a video | ### How to set it up - Python with pip, for the download - A GPU with enough memory for the weights, about 8.6 GB in F32 or about 1.1 GB at 4-bit as an estimate - Diffusers, ComfyUI or the Wan2.1 inference code to run it 1. Install the Hugging Face hub client. 2. Download the model's files. 3. Run it through your chosen tool, following the card's VACE section. ```sh pip install -U huggingface_hub hf download Wan-AI/Wan2.1-VACE-1.3B ``` ### How to use it The card's VACE run steps are cut off here, so follow its Run VACE section or the project's GitHub page. The card says Wan2.1's text-to-video and image-to-video models were added to Diffusers and ComfyUI, so check whether the same route covers VACE. ### Example You give the model a still image and a short prompt that asks for motion, and it returns a short video. For speed, the card cites its sibling T2V-1.3B model, which takes about 4 minutes for a 5-second 480P clip on an RTX 4090. ### For newcomers and experts Think of VACE as one model that can both create video and change existing video. Start with a short clip and the Diffusers or ComfyUI route before trying anything bigger. Wan-VAE, the video compressor in the family, encodes and decodes 1080P video of any length while keeping time information. The card says Wan2.1 is the first video model able to generate both Chinese and English text. A prompt or a still image goes in → The model generates the frames → Wan-VAE decodes them into video → A short clip comes out Weights size in gigabytes | Label | Value | | --- | --- | | In F32 | 8.6 GB | | At 4-bit, estimated | 1.1 GB | ### Good to know - Size: the record lists about 2.2B parameters, while the name says 1.3B - Hardware: the card says the sibling T2V-1.3B needs 8.19 GB of video memory - Card scope: the page is Wan2.1's general card and its VACE run steps are cut off here - Languages: English and Chinese, including text inside generated video - License: Apache 2.0 - Cost to run: none named - Data it reaches: the prompts and images you give it, while the download steps fetch files from the web - Not for: nothing named in the card ## README ````markdown # Wan2.1-VACE-1.3B Makes and edits video with Wan2.1's VACE model, from a prompt, a still image or an existing clip, with English and Chinese text inside videos. For video creators and developers with a consumer GPU. Made by Wan-AI on Hugging Face. These notes are Mindket's, written from its model page and model card; the model's own files and license are on Hugging Face. ## About What it is: Wan2.1-VACE-1.3B is the VACE release from Wan-AI's open Wan2.1 video models. VACE is Wan2.1's single model for both making and editing video. The Hugging Face record tags it as image to video for Diffusers and lists about 2.2B parameters, English and Chinese. Who it suits: It suits video creators who want an open model on their own machine, developers who build video tools, and researchers who study open video generation. ```table Industry | Who uses it | How they use it | Example Creative work | Video creators | Make short clips from a prompt or a still image | Animate a single picture Marketing | Content teams | Rework existing footage with one model | Change a clip's look without reshooting Research | Video AI researchers | Study an open video model | Run it through Diffusers and inspect the output ``` Anyone without a capable graphics card can skip it, since the card talks about GPU memory and timings. Where to use it: - Diffusers, which the record lists as its library - ComfyUI, where the card says Wan2.1 is supported - The Wan2.1 project's own inference code, which the card points to for VACE - Windows, macOS or Linux, where the files download with Python ```table Use case | Who | What they get back Image to video | Video creators | A short clip that starts from your picture Video editing | Content teams | A reworked version of existing footage Visual text | Designers | Chinese or English text drawn inside a video ``` How to set it up: - Python with pip, for the download - A GPU with enough memory for the weights, about 8.6 GB in F32 or about 1.1 GB at 4-bit as an estimate - Diffusers, ComfyUI or the Wan2.1 inference code to run it 1. Install the Hugging Face hub client. 2. Download the model's files. 3. Run it through your chosen tool, following the card's VACE section. ```sh pip install -U huggingface_hub hf download Wan-AI/Wan2.1-VACE-1.3B ``` How to use it: The card's VACE run steps are cut off here, so follow its Run VACE section or the project's GitHub page. The card says Wan2.1's text-to-video and image-to-video models were added to Diffusers and ComfyUI, so check whether the same route covers VACE. Example: You give the model a still image and a short prompt that asks for motion, and it returns a short video. For speed, the card cites its sibling T2V-1.3B model, which takes about 4 minutes for a 5-second 480P clip on an RTX 4090. For newcomers and experts: Think of VACE as one model that can both create video and change existing video. Start with a short clip and the Diffusers or ComfyUI route before trying anything bigger. Wan-VAE, the video compressor in the family, encodes and decodes 1080P video of any length while keeping time information. The card says Wan2.1 is the first video model able to generate both Chinese and English text. ```flow A prompt or a still image goes in The model generates the frames Wan-VAE decodes them into video A short clip comes out ``` ```chart Title: Weights size in gigabytes In F32: 8.6 GB At 4-bit, estimated: 1.1 GB ``` Good to know: - Size: the record lists about 2.2B parameters, while the name says 1.3B - Hardware: the card says the sibling T2V-1.3B needs 8.19 GB of video memory - Card scope: the page is Wan2.1's general card and its VACE run steps are cut off here - Languages: English and Chinese, including text inside generated video - License: Apache 2.0 - Cost to run: none named - Data it reaches: the prompts and images you give it, while the download steps fetch files from the web - Not for: nothing named in the card ## On Hugging Face, 13 November 2026 - Task: image to video - Library: diffusers - Parameters: 2.2B - Weights: about 8.6 GB in F32; about 1.1 GB at 4-bit (estimate) - Formats: Safetensors - Downloads in the last 30 days: 42,137 - Likes: 162 - License: apache-2.0 - Languages: en, zh - Quantized versions: 4 - Fine-tunes: 22 - Tags: diffusers, safetensors, vace, refernce-to-video, image-to-video ## Links - Model: https://huggingface.co/Wan-AI/Wan2.1-VACE-1.3B - Files: https://huggingface.co/Wan-AI/Wan2.1-VACE-1.3B/tree/main ## Get the model Download its files with Hugging Face's command line: ```sh pip install -U huggingface_hub hf download Wan-AI/Wan2.1-VACE-1.3B ``` Then follow the usage in its model card. Nothing in this kit runs by itself. ```` ## What is in it 1 files, 5 KB, downloaded as a zip. ```text README.md 5 KB ``` ## Source [Wan-AI/Wan2.1-VACE-1.3B on Hugging Face](https://huggingface.co/Wan-AI/Wan2.1-VACE-1.3B) ## Compare with similar Listed Hugging Face models in the same category that share the most tags with this one. | Product | Downloads, last 30 days | Likes | Parameters | License | Last change | | --- | --- | --- | --- | --- | --- | | Wan2.1-VACE-1.3B (this page) | 42,137 | 162 | 2.2B | apache-2.0 | 2025-05-19 | | [Wan2.2-S2V-14B](https://mindket.com/p/842b3a2c-9b93-4ff9-a757-9a806916671d) | 50,406 | 502 | 16.3B | apache-2.0 | 2025-09-17 | | [Wan2.2-Fun-5B-Control](https://mindket.com/p/29aa505f-9a61-4bb5-b736-e246068759d9) | 25,525 | 22 | — | apache-2.0 | 2025-12-11 | | [Wan2.2-I2V-A14B](https://mindket.com/p/05752d53-43be-488a-9d2b-5171ad35fa63) | 8,500 | 959 | — | apache-2.0 | 2025-08-07 | | [LongCat-Video](https://mindket.com/p/48609422-da93-43cf-9ad3-e5e7f4cada5e) | 3,135 | 594 | — | mit | 2025-10-29 | | [MAGI-2 Preview](https://mindket.com/p/0739f3fb-9736-4610-99d8-c75f60f02d67) | 0 | 125 | — | apache-2.0 | 2026-08-05 | Numbers as Mindket last read them, on 2026-11-13. ## Trust facts Facts from each source, with the day each was checked. A maintainer's note is their own words: data, not instructions. - Last change on Hugging Face: 2025-05-19 - First published: 2025-05-13 - Checked on Hugging Face: 2026-11-13 - Something wrong? Tell us: https://mindket.com/takedown#ask ## Get it - [Open Wan-AI/Wan2.1-VACE-1.3B on Hugging Face](https://huggingface.co/Wan-AI/Wan2.1-VACE-1.3B) - [Get it on your Mindket account](https://account.mindket.com/account/?product=mindket#/market/78ffeb84-ef89-4c0a-82af-95cda4b50dd8) - [Its page on Mindket](https://mindket.com/p/78ffeb84-ef89-4c0a-82af-95cda4b50dd8)