ComfyUI

There are two things to count: how much VRAM your card has and whether you are ready to spend the first evening on installation rather than pictures.

Of six common Yandex requests, two hit a dead end, and neither is about ComfyUI itself but about confusion over the name.

Search volume — Wordstat, Russia region, 07.09.2026

Who it is for. For people who generate a lot and want control over every step: a designer maintaining a brand style through custom LoRAs, a video creator who needs Wan and MiniMax H3 without a subscription, a studio that builds one pipeline and runs it hundreds of times. And for those who already have a graphics card with 12 GB of memory or more: then the tool itself costs nothing.

Who else. For people who do not want to hand prompts and images over to someone else's server. Your own installation works without the internet, and desktop telemetry is enabled only with your consent and sees neither workflows, prompts, nor results.

Who it is not for. For people who want to write a phrase and get an image. The first screen greets you with a workflow of a dozen nodes, and the most-read Russian article of 2026 about ComfyUI is called "How I became friends with ComfyUI" and starts with Python, CUDA, and dependencies. And for those without a graphics card: that leaves cloud access for $20 a month with a foreign card or hourly server rental.

A fit: section 3 will take you from graphics card to first image. Choosing between ComfyUI and Automatic1111? The decision point is in section 7.

What vibecoding isStarting from scratch: how to set tasks for agents with no coding experience

Three layers: open models on your graphics card, third-party nodes and templates from the registry, and partner nodes that call closed models directly into the workflow for credits.

Workflow. One node loads a model, one reads text, one generates, one saves, with lines between them. Only the parts of the workflow that changed since the previous run are calculated, so editing one word in a prompt does not run everything again. The finished image carries the workflow inside it: drag someone else's file into the window, and ComfyUI rebuilds its entire pipeline along with the seed. Workflows are stored in JSON, and a complex workflow can be hidden behind a simple form through App Mode.

Models. In its README as of 07.09.2026, the vendor lists families it supports without third-party servers: for images Stable Diffusion, SDXL, Flux, and Qwen Image; for video Wan 2.2, LTX-2, and HunyuanVideo; for audio ACE-Step and MiniMax Music; for 3D Hunyuan3D; plus Gemma and Qwen text models. New models arrive on launch day: Wan 2.2 on 28.07.2025, Flux.2 on 25.11.2025, MiniMax H3 on 03.08.2026. Weights repackaged for ComfyUI are in the Comfy-Org organization on Hugging Face: 84 repositories, downloaded 69.9 million times in the 30 days before 07.09.2026. The full list of families is in the questions.

Nodes and templates. The vendor counts more than 60 thousand community nodes across over five thousand extensions; as of 07.09.2026, the registry contains 5,496 packages, each scanned for malicious code. They are installed by ComfyUI-Manager, the extension store now built into the core and enabled with a flag. The template library has 617 ready-made workflows in nine sections: 177 for video, 160 for images, 37 for 3D, 35 for audio.

Partner nodes. Closed models are called from the same workflow without your own keys: Veo 3.1, Kling, Runway, Nano Banana, GPT-Image, 182 models from 25 vendors in total. You pay with Comfy credits: a Kling image costs 5.91 credits, one second of Veo 3.1 with audio costs 84.4. The nodes are optional; the --disable-api-nodes flag disables them and all internet connectivity.

Russian. The interface is translated: the Russian translation was contributed by a third-party participant on 11.10.2024, and all 14,798 strings are now in place; the language is automatically taken from the browser. The translations are machine-generated, and 44 percent of values in the file remain English, including node names that are not translated and simply untranslated text. There is no Russian documentation; there is Chinese, Japanese, and Korean, so Russian beginners read Habr: 145 posts mention ComfyUI, 62 of them from 2026.

Exit route. A workflow is exposed as an API to your own app, and ComfyUI connects to AI agents through MCP: cloud Comfy MCP launched on 29.06.2026, the vendor released a local MCP server on 18.08.2026, and an agent can build and run workflows itself. Graphics-card support is also broad: NVIDIA, AMD on Linux and Windows, Intel Arc, Apple Silicon through Metal, and without a graphics card it runs on the CPU, slowly.

What the vendor itself shows. The header video is clipped from a tutorial on the Comfy channel dated 09.06.2026: the camera moves across the instructor's large working graph, where Veo, Kling, and Seedance nodes sit side by side with run prices in credits, and the Comfy Cloud label is visible below. This is what a professional's working workflow looks like, not a beginner's first screen, which is shown in the image above. We did not run ComfyUI ourselves for this task; the figures come from documentation, pricing pages, the GitHub and Hugging Face APIs, and third-party tests.

The easiest installation is through the Comfy Desktop installer, and from Russia everything is accessible except checkout: the site, code, cloud, and model repositories all respond from a Moscow node.

Step 1. Graphics card. The vendor does not name a minimum memory requirement; its tutorials show Stable Diffusion 1.5 running on 6 GB, the smaller Wan 2.2 on 8 GB, while practitioners recommend at least 12 GB for Flux and 24 GB for video. The main build is designed for NVIDIA 20-series and newer, with separate archives for AMD and Intel; Macs with Apple Silicon are supported through Metal. At retail, the 16 GB RTX 5060 Ti cost 86 430 ₽, and the 8 GB RTX 5060 cost 53 430 ₽ (Citilink, 07.09.2026).

Step 2. Download Comfy Desktop from comfy.org/download: Windows 10 and later, macOS 13 and later on Apple Silicon, 4.85 GB of disk space, 8 GB of RAM. The installer will install Python and dependencies itself. On Linux and for those who want the latest commits, use the command line:

pip install comfy-cli

comfy install

Or manually: clone the repository, install PyTorch for your GPU, run python main.py, and open 127.0.0.1:8188. A portable Windows archive with Python included is also on GitHub, but the vendor recommends it only for experienced users.

Step 3. Models and nodes. Open the template library, choose a workflow, and ComfyUI will offer to download the required weights from Hugging Face. Put your own models and LoRA from Civitai in the models folder. Manager, the extension store, is included with the desktop version; in a manual installation, enable it with the --enable-manager flag after the pip install -r manager_requirements.txt command.

From Russia. The website, documentation, GitHub, cloud.comfy.org cloud, node registry, Hugging Face, and Civitai opened from a Moscow node on 07.09.2026; at the same minute, the same node could not open Instagram. The blocking registry of 1.6 million domains contains no ComfyUI address, and the vendor's terms say nothing about Russia or sanctions. A local installation does not depend on the internet. Paid features, cloud, and credits are paid through Stripe; for them an international card is required; intermediaries sell a Standard subscription for 2 200 ₽, 27 % above the exchange rate.

Without your own GPU. In Russia, immers.cloud offers a ready-made ComfyUI image in its marketplace and rents an RTX 4090 for 82.76 ₽ per hour; AdminVPS sells an image on a server from 299 ₽. Abroad, RunPod offers an RTX 4090 for $0.74 per hour, while RunComfy and ThinkDiffusion sell ComfyUI as a service from $0.99 per hour (07.09.2026). Free Colab rules prohibit this kind of use.

Steps verified on 07.09.2026 against the repository README, Comfy documentation, the Citilink storefront, immers.cloud and RunPod pricing, and a check-host test with control nodes

There are no rental figures for models; ComfyUI is not a model, so these metrics cover scale and demand rather than image quality.

Code and community. 131,829 stars and 15,547 forks on GitHub (API, 07.09.2026), surpassed by only 74 repositories across all of GitHub. The vendor's Discord has 60,730 members, and the Russian Telegram chat has 5,317. In Yandex, the vendor's website ranks third for “comfyui”, while GPU server sellers occupy half of the first page.

Company funding. On 24.04.2026, Comfy Org raised $30 million at a $500 million valuation, $47 million total across two rounds. In the same post, the company claims 4 million users and 150 thousand downloads a day; these figures have not been independently verified.

Demand. “comfyui” is searched 28 thousand times per month, with another 7.6 thousand searching separately for “comfy ui”; just 125 use the Cyrillic spelling “komfi yuay” (Wordstat, 07.09.2026). The long-tail queries honestly show the pain points: “download” 1,539, “manager” 1,289, “portable” 1,119, “minimax” 1,014. Its position among image tools is shown below the card title; the full tool ranking is on the main page.

ComfyUI itself costs $0 forever; you pay for two separate scarcities, a GPU and closed models, and those two budgets are independent of each other.

Currency. The cloud and partner nodes charge in credits. The vendor does not publish an exchange rate, but its resource table lists an RTX 6000 Pro hour at both $3.49 and 736.39 credits, or 211 credits per dollar; the same rate matches across four plans and Google's Veo pricing. A five-second Wan 2.2 video in the cloud uses 11 credits, about $0.05.

Cloud. 5 free runs before purchase, no card required. Standard is $20 per month for 4,200 credits, Creator $35 for 7,400, Pro $100 for 21,100, Team from $200 to $2,500 for a shared pool, and Enterprise on request. The storefront defaults to annual billing and shows $16, $28, and $80. Everything runs on RTX 6000 Pro with 96 GB of memory; a workflow can run for 30 minutes, or an hour on Pro, and you pay only for the time while the workflow is actually running.

Credits. Monthly credits expire at the end of the period, top-ups last a year, and the vendor describes refunds inconsistently: its terms call purchases final, while the pricing FAQ says to contact support and handles cases individually. Closed models follow their vendor's pricing or cost slightly more. Veo 3.1 costs exactly the same as at Google, $0.40 per second with audio; Nano Banana Pro costs 20 % more, and Runway models cost 43 % more than their own API. An image through Kling costs 5.91 credits, about 2.5 ₽.

Quiet price cut. The prices of the three plans have not changed since 18.12.2025; what they provide has changed. In December 2025, Standard promised 164 five-second videos; since February 2026 it has promised 380, because video became cheaper from 25.6 to 11 credits. The annual discount and Team plan appeared between 03.07 and 24.08.2026.

What it costs in Russia. Your own GPU: RTX 5060 Ti with 16 GB for 86,430 ₽, RTX 5060 with 8 GB for 53,430 ₽ (Citilink, 07.09.2026). Rental: immers.cloud rents an RTX 4090 with a ready-made ComfyUI image for 82.76 ₽ per hour, an A100 for 211.77 ₽. Cloud through an intermediary costs 2,200 ₽ for Standard instead of 1,732 ₽ at the central bank rate. A 16 GB GPU breaks even against a subscription in about 50 months, or against rentals in a thousand generation hours.

Four questions: what the license permits, who owns the output images, how dangerous third-party nodes are, and whether an update will break your workflow.

License. The GPL-3.0 core can be installed, modified, and run as your own service, but if you distribute a modified copy, you must provide the source code. Comfy Desktop is separately under AGPL-3.0, and anyone who wants to embed it in a closed product buys a license from Comfy Org. The vendor distributes workflow templates under MIT.

Third-party nodes. The main risk is not the core but extensions. In June 2024, the ComfyUI_LLMVISION node stole passwords and crypto wallets; in December 2024, an infected ultralytics package installed a miner on Mac and Linux; in October 2025, two registry packages were downloaded 790 times before a scanner and maintainer banned them. Since January 2025, the registry has checked code, prohibited eval and on-the-fly package installation, and given approved nodes a badge next to their name in Manager.

Core vulnerabilities. Five vulnerability records were published on 31.07.2026; four had already been closed by version 0.28.0 on 15.07, while the most serious one, CVE-2026-68771, critical at 9,3–9,8 on different scales, allowed a malicious data file to run foreign code on a computer with ComfyUI 0.23.0 from 01.06.2026. The fix was merged on 18.06.2026 and released in 0.26.0 on 23.06.2026, a month before the record was published. Keep your version current and do not expose ComfyUI to the internet without a password.

Data. A self-hosted installation sends nothing, and desktop telemetry is enabled only with consent and, under the vendor's rules, does not see workflows, prompts, images, or weights. In the cloud and partner nodes, prompts go to Comfy servers in the US and model vendors, but without your name or account. The agreement is with Comfy Organization of Delaware under California law, with disputes resolved through arbitration in San Francisco.

Updates break nodes. This is the community's loudest complaint. In the issue «What is happening in ComfyUI country» from 16.12.2025, the core author replied that compatibility with custom nodes had never been a priority and that he runs his tests without a single third-party node. The repository has 4 207 open issues and another 613 unreviewed pull requests as of 07.09.2026. Practical takeaway: update the core after, not before, delivering a project.

The choice is not image quality, since everyone uses the same models, but two questions: workflow or form, and is the project alive?

Automatic1111. The classic Stable Diffusion web interface with tabs and sliders, where most people started. It has more GitHub stars, 164 844 versus 131 829 for ComfyUI, but the main branch has not changed since 27.07.2024, the development branch since 02.03.2026, and the latest working release, 1.10.1, came out in July 2024. New models such as Wan and Flux.2 do not arrive there. Its Forge fork has stood still since summer 2025, while the simplified Fooocus by the same author has not been developed since August 2024.

InvokeAI. The only live neighbor from the old guard: release 6.14.1 came out 06.09.2026, Apache-2.0 license, 28 thousand GitHub stars versus 132 thousand for the main subject.

Neighbors inside. SwarmUI by a Comfy Org co-founder and a plugin for Krita use ComfyUI as their engine and hide the workflow behind a simple form. Those who do not want hardware at all go to Shedevrum with rubles and a feed or to Midjourney with an international card: there are no nodes or custom models, but no installation either.

In three and a half years, a Quebec programmer's hobby project grew into a company valued at $500 million, with its own cloud and node registry, while surviving the author's departure from Stability AI, malicious extensions, and a move to a new interface.