Moss, Moss-Image & Moss Turbo LoRAs
MOSS-Image is a 5B open text-to-image model from the OpenMOSS team, released on 16 October 2026. It reads prompts through a Qwen3.5 9B text encoder, uses the FLUX.2 VAE, and is built for text and layout: English and Chinese text rendering, long prompts and JSON layout boxes. It takes LoRAs, and Moss Turbo runs it in 8 steps with a community-made LoRA.
Keep every Moss, MOSS-Image and Moss Turbo LoRA in your private vault.
Plans from 100 GB to 5 TB. From $4.99 a month.
Keep your Moss LoRAs in one private vault
Drop your Moss LoRAs into BigVault and they sit in our own encrypted storage, next to the rest of your models. Only you can see your vault. Keep one folder per checkpoint so the right file is always at hand:
loras/moss-base/for LoRAs made on MOSS-Image Baseloras/moss-sft/for LoRAs made on SFTloras/moss-turbo/for the 8-step Turbo LoRA
Add a Moss LoRA from Civitai in one click
Paste a Civitai model link in your vault and BigVault adds it in one click. It keeps the model's details and trigger words with the file, and tidies file names and descriptions. Imports run on our servers in the background, so you can close the page.
Building a collection? Read how to keep your own Civitai archive.
Render Moss on Graydient (Beta)
Graydient has a Moss workflow for MOSS-Image. Rendering a model from your vault on Graydient is in Beta and needs a Graydient plan; .safetensors files under 25 GB are compatible, apart from a few edge cases. Pasting vault links into the Graydient dashboard is coming soon.
Graydient: “Use 100,000+ Civitai models, instantly installable in one click.”
Base, SFT or Turbo: which one your LoRA needs
MOSS-Image comes as two checkpoints, and Moss Turbo is a third way to run it.
MOSS-Image Base (most LoRAs)
Base is the starting point most Moss LoRAs are made on.
MOSS-Image SFT (Base LoRAs work)
SFT gives cleaner layouts. LoRAs made on Base also load on SFT. You need ComfyUI 0.42.0 or later for either.
Moss Turbo (community 8-step LoRA)
Moss Turbo runs MOSS-Image in 8 steps with a community-made 8-step Turbo LoRA. You keep it like any other LoRA; its size varies by version.
How big are Moss LoRAs?
In a sample of 40 Moss LoRAs on Civitai (public Civitai API, 22 October 2026) the median file was 138.4 MB, and sizes ran from 8.9 MB to 1,080 MB. They are .safetensors files.
| Plan | Storage | Price | About how many fit (138 MB median) |
|---|---|---|---|
| Locker | 100 GB | $4.99 a month | ≈ 722 |
| Library | 500 GB | $14.99 a month | ≈ 3,612 |
| Archive | 1 TB | $24.99 a month | ≈ 7,225 |
| Studio | 5 TB | $99 a month | ≈ 36,127 |
Moss LoRAs to start with
Moss LoRAs are just arriving. Keep the ones you find in your vault: paste a Civitai link and BigVault adds it in one click.
Moss LoRA questions
- Does MOSS-Image support LoRAs?
- Yes. LoRAs are usually made on MOSS-Image Base, and they also load on SFT.
- What is Moss Turbo?
- An 8-step way to run MOSS-Image using a community-made Turbo LoRA. You keep it like any other LoRA.
- Which ComfyUI version do I need?
- 0.42.0 or later.
- How big is a Moss LoRA?
- In a sample of 40 Moss LoRAs on Civitai the median was about 138 MB.
- Where do Moss LoRAs go?
- In ComfyUI, in ComfyUI/models/loras/. One subfolder per checkpoint (moss-base, moss-sft, moss-turbo) keeps Base, SFT and Turbo files apart.
Keep every Moss LoRA in one private place.
Plans from 100 GB to 5 TB. From $4.99 a month.
Questions? info@modeltrends.ai