ComfyUI
Widely used free app for running open image and video models on your own computer; expect a learning curve.
Should you use it?
Pick it if
People who want to make images or video privately on their own GPU, for free, and don't mind some setup.
Best for: Trying new open image and video models early, on your own hardware (Run AI on your computer Top 20)
Skip it if
You want the best pictures with no setup; a cloud model like Nano Banana 2.1 is easier.
You give up: Node graph takes time to learn; exposed servers were attacked, keep it private
The short version
- What this is
- A free, open-source app for running image, video, audio and 3D models on your own computer, built around a graph of connected steps (nodes).
- Why it matters
- It is a common way to run open image and video models locally, at no cost and without sending your files anywhere.
- Free GPL-3.0 app for Windows, Linux and macOS, with a desktop installer for Windows and macOS s1.
- Supports most open image and video models, including Qwen Image, Flux and Wan s1.
- Paid Comfy Cloud and partner API nodes are optional; --offline keeps everything local s1.
Good at
- Runs a wide range of open models: image (Flux, Qwen Image and others), video (Wan, LTX-Video, HunyuanVideo), audio (ACE-Step) and 3D s1.
- Works on Windows, Linux and macOS with NVIDIA, AMD, Intel and Apple Silicon hardware s1.
- Very widely used: about 137k GitHub stars, with a weekly release cycle s1; its frontend package was updated on 8 Oct 2026 s2.
Weak at
- Its node graph takes time to learn compared with a simple prompt box s1.
- No independent test of the app itself yet.
Our take
ComfyUI runs a wide range of open image and video models at home. It lists support for Flux, Qwen Image, Wan, LTX-Video, HunyuanVideo, ACE-Step and more s1. The project says large models can run on as little as 4GB of video memory and 8GB of RAM s1, though it will be slow on small machines.
The trade-off is the interface. You build a workflow by connecting nodes, which is powerful but slower to learn than typing a prompt into a cloud app. The desktop installer for Windows and macOS is the easiest start s1. It costs nothing to run locally; Comfy Cloud and the partner API nodes for closed models are the paid extras, and the --offline flag switches those off s1.
We have not scored it because no recent independent test of the app exists. The quality you get depends on the model you load: Qwen-Image 2.1, which you can run in it, scores 48 for images on our page.
Not scored yet
This is a First look: the call above comes from the maker's own pages, so there is no score yet. When a recent independent test or our own run shows how well it does the job on current models.
ImagesVideo
New to this? What the scores and labels mean
- 0 to 100
- Our score for one job, like writing or coding, compared with everything else out today. 100 is the best available today. For AI models we build it from public data, such as Arena's blind votes and Epoch AI's independent tests; apps and tools are scored on our own checklist. The same tool can score differently for different jobs, and where two likely ranges overlap, we call it a tie.
- Our verdict
- Use it: a top choice for most people. Worth trying: good, and worth a go. Depends: good for some people or one kind of job. Wait: too early to tell, or not out for everyone yet. Skip: something else does the job better.
- First look
- Just released. We say what it is and whether to bother, and give it a score once public results exist.
- No score yet
- None of the public sources we use covers it yet, or the only evidence is from a source whose numbers we can't use, so we describe it in words instead.
- Best buy
- Nothing cheaper scores as well for that job.
- Best free
- The best score for that job that costs nothing, through a free plan or by running on your own computer.
- Pay as you go
- No monthly plan: you pay for each use, usually through a developer account.
- Checked
- The day we last confirmed the score and the price against the sources. Prices change often, so check the date.
- AI model
- The engine underneath, like GPT or Claude. You use it through an app or, if you code, directly.
- Open source
- Free to download and run on your own computer, so nothing leaves your machine.
Alternatives
Better or cheaper for images, from our scores, then the ones we'd also look at. The same kind of product comes first.
- GPT Image 2.5 SunburstTop scoredImage model100Pay as you goScores 100 in images, first in our ranking; price not checked. Side by side
- Qwen-Image-2.1Image model48FreeThe top open image model on arena.ai; you can run it inside ComfyUI. Side by side
- Nano Banana 2.1Image model73FreeHigher-ranked cloud image model with a free tier; nothing to install. Side by side
Evidence
Everything behind the call and the scores, folded so the page stays short. Spotted a wrong price, score or link? Report a mistake.
How we scored this
Reviews are drafted with help from AI models, including Anthropic's Claude. Model scores come only from outside leaderboards and independent evaluations, by a written rule.
No one paid for this review or gave us free access, and we earn nothing if you sign up. How we score
Sources 2
Something here out of date or wrong? Report a mistake and we will check it against the source.
History 2 changes
Every change to a rating, the price or our call, with the reason. Scores are against what exists on the day, so they drift down as better things ship. Older entries come from before we moved to scores.
- First look. No recent independent test of the app.
- Audit: softened unsourced claims that it is the standard or usual app.
Use it well 4 guides on this site
Pages on this site that help with the jobs above.