Diffus, 70k models, ComfyUI in the browser
Diffus (diffus.me) is a browser image desk for community checkpoints — not DiffusionBee’s one-click local Mac / Windows Stable Diffusion app, and not FASHN’s garment-on-model try-on. Public pages: pick from 70,000+ hosted models (they split the library as 10,000+ checkpoints and 55,000+ LoRAs), write a prompt, optional ControlNet and inpaint. They also host ComfyUI in a tab: 62 custom node packs and 3,300+ nodes preinstalled, plus Civitai URL import. Treat the output size as the checkpoint / graph on Diffus, not a first-party Diffus 4K picker of this site’s. This page is stills; Diffus also lists video, which is not this card. ImgToVid does not run Diffus; the generator on this page drafts on GPT Image 2.5 Flare.
01 — Capabilities
A hosted ComfyUI cloud, not a local SD installer or a try-on desk
Public Diffus pages sell a Civitai-style model picker and a node graph on their GPUs. They do not sell DiffusionBee’s offline installer, or FASHN’s product-plus-model try-on.
01
A prompt + hosted checkpoint / LoRA → a still on diffus.me
That file stays on Diffus.
02
70,000+ models listed; ComfyUI with 62 packs / 3,300+ nodes
This page does not open that studio.
03
ControlNet, inpaint, ADetailer — not a 14-ref picker here
Not this site’s Diffus generate table.
04
Not DiffusionBee’s local 1024 app, not FASHN Try-On Max 1K / 2K / 4K
DiffusionBee is the offline installer. FASHN is try-on. Diffus is the browser model cloud. This site does not open Diffus.
02 — Prompt to output
Prompts that match the stills above
Each prompt describes the picture on its card. “Use this prompt” sends that text to Create. The generator on this page runs GPT Image 2.5 Flare, not Diffus.
01 · PortraitPromptClose photographic portrait of a young man with curly hair wearing a grey knit sweater, soft window light, he glances aside and smiles warmly, shallow depth of field.
02 · ProductPromptA clean white leather sneaker on a pale grey seamless studio backdrop, soft diffused light, three-quarter view, crisp stitching, minimal commercial photograph.
03 · StreetPromptPhotographic view down a cobblestone lane in an old European village, climbing roses on stone walls, café tables in morning sun, a few people strolling in the distance.
04 · CharacterPromptPhotograph of a young woman with shoulder-length brown hair in an olive t-shirt, golden late-afternoon light, she turns her head and looks into the distance, breeze in her hair.
03 — Where it fits
Jobs a hosted ComfyUI cloud is selling
Public pages point at a community checkpoint plus a prompt, or a graph JSON. They do not sell a DiffusionBee local run, or a FASHN try-on pair.
One still from a listed checkpoint / LoRA, or a ComfyUI graph
That job is Diffus’s generate table. This page does not open that job.
ControlNet / inpaint / ADetailer on their GPUs
Delivery stays on Diffus.
A local DiffusionBee 1024 still, or FASHN garment-on-model
That is DiffusionBee or FASHN. Diffus is the browser model cloud.
A draft still on ImgToVid
Write the still here. GPT Image 2.5 Flare carries the prompt into Create. Membership and credits are checked there.
04 — Verdict
What Diffus actually is
Strengths
- diffus.me: browser SD / SDXL / Flux desk, 70,000+ hosted models
- ComfyUI listed with 62 node packs and 3,300+ nodes
- ControlNet, inpaint, ADetailer, Civitai URL import named
- Honest that size follows the checkpoint / graph, not a 4K Nano Banana picker here
Trade-offs
- No first-party Diffus 4K generate picker here — output follows the model they host
- A Diffus job on diffus.me is not this site’s generate table
- Diffus is not available on ImgToVid
- Marked coming soon on this site
05 — Side by side
Diffus, Nano Banana 2 and GPT Image 2.5
Diffus is a hosted ComfyUI / checkpoint cloud. Nano Banana 2 and GPT Image 2.5 numbers below are from those vendors’ public docs on ImgToVid. A 70k-model still there is not a Diffus generate row here.
06 — How it works
Start a still on ImgToVid
Step 1
Describe one still
Subject, light, frame. Diffus will not be the model that runs it.
Step 2
Use the generator on this page
It runs GPT Image 2.5 Flare, because Diffus is not offered here. The prompt and the settings carry into Create.
Step 3
Generate in Create
Membership and credits are checked there. Coming back to this page does not switch the model over to Diffus.
07 — FAQ
Diffus, in plain terms
What is Diffus?
Diffus is a browser Stable Diffusion / Flux desk on diffus.me, with 70,000+ hosted models and an online ComfyUI. ImgToVid does not host Diffus.
What resolution does Diffus output?
Size follows the checkpoint or ComfyUI graph they host. That is not a Diffus 4K picker here.
Does Diffus generate images from text?
Yes — a prompt plus a hosted model, optional ControlNet. That is diffus.me, not a one-shot still on ImgToVid.
How many images per run?
The public simple generator returns one still. This site does not open that desk.
Can I generate Diffus images on ImgToVid?
No. The generator on this page runs GPT Image 2.5 Flare so you can draft a still. Diffus is marked coming soon.
Can I use the results commercially?
Commercial use of anything you generate on ImgToVid depends on your plan. Check Pricing and our Terms before you publish, and make sure prompts and uploads do not infringe someone else’s rights.
08 — Explore more
Other AI image models
Diffus is not the button on this page
You can still draft the still. The generator runs GPT Image 2.5 Flare, and the prompt goes with you into Create.
Draft in Create