Skip to content
ModelListedExternal
Browse files

Bernini 1.3B

> 💬 Discord — updates, roadmaps, projects, or just to chat > > 💻 Code / ComfyUI nodes > > 🧬 Bernini-R 1.3B > > 14B GGUF (full quality)

Sourcehuggingface.co/neuregex/Bernini-1.3B-ComfyUI/tree/a34c3f96f60566e47b8729712ac92a29f87955b1
KindText encoderBase modelByteDance/Bernini-R-1.3B-DiffusersVersionRevision a34c3f9Bgrade

Model source

Kind
Text encoder, UNet, VAE
File size
242 MB–6.27 GB
Base model
ByteDance/Bernini-R-1.3B-Diffusers
Version
vRevision a34c3f9
Updated
Oct 2, 2026
Downloads
2K
Tasks
Image edit, Text to video, Video edit
Source
Hugging Face

Source description

Bernini-R 1.3B — ComfyUI bundle (everything you need)

💬 Discord — updates, roadmaps, projects, or just to chat

💻 Code / ComfyUI nodes

🧬 Bernini-R 1.3B

14B GGUF (full quality)

The small, accessible Bernini-R for ComfyUI. This repo is self-contained — the 1.3B renderer plus the VAE and text encoder it needs — so you can edit images/video on modest GPUs. It's the lightweight sibling of the 14B GGUF repo (that one is full quality for 24 GB cards).

Bernini-R 1.3B is single-expert (fine-tuned from Wan2.1-T2V-1.3B — no high/low MoE), so it's tiny (~2.6 GB) and runs almost anywhere. It performs close to the 14B on simpler edits (style transfer, watermark/subtitle removal, local editing) and lags on the hardest tasks.

Files

FileWhatSizeFolder in ComfyUI
bernini_r_1.3B-bf16.safetensorsrenderer (native, full precision)~2.8 GBmodels/diffusion_models/
bernini_r_1.3B-Q8_0.ggufrenderer (GGUF, near-bf16)~1.5 GBmodels/unet/
bernini_r_1.3B-Q6_K.ggufrenderer (GGUF, high quality)~1.1 GBmodels/unet/
bernini_r_1.3B-Q5_K_M.ggufrenderer (GGUF, balanced)~1.0 GBmodels/unet/
bernini_r_1.3B-Q4_K_M.ggufrenderer (GGUF, smallest)~0.8 GBmodels/unet/
vae/wan_2.1_vae.safetensorsWan 2.1 VAE~0.25 GBmodels/vae/
text_encoders/umt5-xxl-encoder-Q5_K_M.ggufUMT5 text encoder (GGUF — used by the workflows)~4.1 GBmodels/text_encoders/
text_encoders/umt5_xxl_fp8_e4m3fn_scaled.safetensorsUMT5 text encoder (fp8 — optional fallback)~6.4 GBmodels/text_encoders/

Usage in ComfyUI

  1. Install ComfyUI-BerniniR and ComfyUI-GGUF (required — the graphs load the text encoder with CLIPLoaderGGUF).
  2. Drop each file in the folder from the table above.
  3. Load the workflow workflows/bernini_i2i_1.3B.json (or grab it from the node repo). It's a single-expert graph — no model_low. Put your image, run.
    • Renderer — native (bf16): BerniniR · Load Model (native). GGUF: UnetLoaderGGUF → BerniniR · Apply Patches → Source Stream → Guider (leave model_low empty).
    • Text encoder: CLIPLoaderGGUF (type = wan) → umt5-xxl-encoder-Q5_K_M.gguf. The graphs default to the GGUF encoder because the fp8 .safetensors one triggers a Windows / torch-2.8 access violation under memory pressure; the fp8 file is kept only as an optional fallback.

UMT5 GGUF text encoder quantized by city96.

Need full quality on 24 GB?

Use the 14B dual-expert GGUFs: neuregex/Bernini-R-GGUF. Same nodes, just wire both experts into the guider (model = high, model_low = low).

License: Apache-2.0 (same as Bernini-R and the Wan base).

Sources

1 source
Hugging FacePrimaryVersion a34c3f96f60566e47b8729712ac92a29f87955b1
huggingface.co/neuregex/Bernini-1.3B-ComfyUI/tree/a34c3f96f60566e47b8729712ac92a29f87955b1

Verified Oct 2

Download files

Model artifacts

8 artifacts

bernini_r_1.3B-bf16.safetensors

safetensors · 2.64 GB · SHA-256 d6d21e650af9…9819 · Hugging Face

Download file

bernini_r_1.3B-Q4_K_M.gguf

gguf · 937 MB · SHA-256 ce68d520a3b5…b485 · Hugging Face

Download file

bernini_r_1.3B-Q5_K_M.gguf

gguf · 1.01 GB · SHA-256 b9027f702b13…5ff4 · Hugging Face

Download file

bernini_r_1.3B-Q6_K.gguf

gguf · 1.12 GB · SHA-256 0e37faeec6ad…68f8 · Hugging Face

Download file

bernini_r_1.3B-Q8_0.gguf

gguf · 1.43 GB · SHA-256 f8b623d01bd3…4c82 · Hugging Face

Download file

text_encoders/umt5_xxl_fp8_e4m3fn_scaled.safetensors

safetensors · 6.27 GB · SHA-256 c3355d30191f…4f68 · Hugging Face

Download file

text_encoders/umt5-xxl-encoder-Q5_K_M.gguf

gguf · 3.86 GB · SHA-256 eaea358bb438…12ce · Hugging Face

Download file

vae/wan_2.1_vae.safetensors

safetensors · 242 MB · SHA-256 2fc39d31359a…976b · Hugging Face

Download file

Source excerpts

3 excerpts
ReadmeJul 16huggingface.co/neuregex/Bernini-1.3B-ComfyUI

--- license: apache-2.0 base_model: ByteDance/Bernini-R-1.3B-Diffusers pipeline_tag: image-text-to-video library_name: gguf tags: - comfyui - bernini-r - wan2.1 - text-to-video - image-editing - video-editing - gguf --- # Bernini-R 1.3B — ComfyUI bundle (everything you need) > **[💬 Discord — updates, roadmaps, projects, or just to chat](https://discord.gg/HxfP9TnctJ)** > > **[💻 Code / ComfyUI nodes](https://github.com/neuregex/ComfyUI-BerniniR)** > > **[🧬 Bernini-R 1.3B](https://huggingface.co/ByteDance/Bernini-R-1.3B-Diffusers)** > > **[14B GGUF (full quality)](https://huggingface.co/neuregex/Bernini-R-GGUF)** The **small, accessible** Bernini-R for ComfyUI. This repo is **self-contained** — the 1.3B renderer **plus** the VAE and text encoder it needs — so you can edit images/video on modest GPUs. It's the lightweight sibling of the [14B GGUF repo](https://huggingface.co/neuregex/Bernini-R-GGUF) (that one is full quality for 24 GB cards). Bernini-R 1.3B is **single-expert** (fine-tuned from Wan2.1-T2V-1.3B — no high/low MoE), so it's tiny (~2.6 GB) and runs almost anywhere. It performs close to the 14B on simpler edits (style transfer, watermark/subtitle removal, local editing) and lags on the hardest tasks. ## Files | File | What | Size | Folder in ComfyUI | |------|------|------|-------------------| | `bernini_r_1.3B-bf16.safetensors` | renderer (native, full precision) | ~2.8 GB | `models/diffusion_models/` | | `bernini_r_1.3B-Q8_0.gguf` | renderer (GGUF, near-bf16) | ~1.5 GB | `models/unet/` | | `bernini_r_1.3B-Q6_K.gguf` | renderer (GGUF, high quality) | ~1.1 GB | `models/unet/` | | `bernini_r_1.3B-Q5_K_M.gguf` | renderer (GGUF, balanced) | ~1.0 GB | `models/unet/` | | `bernini_r_1.3B-Q4_K_M.gguf` | renderer (GGUF, smallest) | ~0.8 GB | `models/unet/` | | `vae/wan_2.1_vae.safetensors` | Wan 2.1 VAE | ~0.25 GB | `models/vae/` | | `text_encoders/umt5-xxl-encoder-Q5_K_M.gguf` | UMT5 text encoder (**GGUF — used by the workflows**) | ~4.1 GB | `models/text_encoders/` | | `text_encoders/umt5_xxl_fp8_e4m3fn_scaled.safetensors` | UMT5 text encoder (fp8 — optional fallback) | ~6.4 GB | `models/text_encoders/` | ## Usage in ComfyUI 1. Install **[ComfyUI-BerniniR](https://github.com/neuregex/ComfyUI-BerniniR)** and **[ComfyUI-GGUF](https://github.com/city96/ComfyUI-GGUF)** (required — the graphs load the text encoder with `CLIPLoaderGGUF`). 2....

NoteOct 2huggingface.co/neuregex/Bernini-1.3B-ComfyUI

Source context: 1976 downloads · 21 likes · Pipeline image-text-to-video · Library gguf · Repo neuregex/Bernini-1.3B-ComfyUI

NoteJul 16huggingface.co/neuregex/Bernini-1.3B-ComfyUI

Source context: 1995 downloads · 15 likes · Pipeline image-text-to-video · Library gguf · Repo neuregex/Bernini-1.3B-ComfyUI