Skip to content
Other

spriteforge-v3-comfy

ComfyUI LoRA adaptation for sprite sheet generation

2 min readupdated

This is a repackaged version of the SpriteForge LoRA by developer tantk7, adapted for the ComfyUI format with modified tensor names. The model builds upon the Qwen-Image-Edit-2511 base and is designed to automatically transfer a character from a reference image into a ready-made sprite sheet template (4x4, 1024x1024). This tool will be useful for game developers, animators, and artists creating sprites for 2D projects.

Key features

  • Character transfer: replaces the character in the sprite sheet template with the hero from the reference.
  • ComfyUI compatibility: weights are renamed for correct operation without extra errors.
  • Fixed format: designed to work with 4x4 sprite grid sizes.
  • Base model: built on Qwen-Image-Edit-2511.

What you can do

  • Generate a character sprite set for a 2D game based on a single reference image.
  • Use a 1024x1024 pixel sprite sheet template to obtain frame-by-frame animation.
  • Combine a character reference and a ready template in ComfyUI following the author's instructions.

Example

Use a prompt like: «Replace the character in the sprite sheet with the character from the reference image» when feeding two images (image1 — character reference, image2 — sprite sheet template).

Read moreDetails
Base model: Qwen-Image-Edit-2511
Weights format: modified for ComfyUI (added diffusion_model. prefix, removed PEFT .default suffix)
License: Apache 2.0
Original: model by tantk7 from ModelScope
Discuss Telegram 0

Will it run?

Runs in FP8 / INT8

Full precision does not fit, a quantised version does: quality barely changes down to 8 bits, more noticeably at 4 bits.

  • BF16weights 38 GB · ready files43 GB
  • FP8 / INT8recommendedweights 19 GB22 GB
  • 4-bitweights 11 GB13 GB

Tip: go 8-bit.FP8 / INT8 is within about a percent of full precision in quality, takes half the memory and runs faster on RTX 40/50 and H100. Full precision only pays off for fine-tuning.

What to get for this model

Massed ComputeOpen Massed Compute

Your own GPU: FP8 / INT8 — from RTX 4090

Massed Compute hourly prices, November 2026. The discounted price applies with the coupon at checkout.

This is a LoRA: counted together with its base model Qwen-Image-Edit-2511.

Estimate: weights plus working memory. Long videos, high resolutions and long contexts need more.

Run it on Massed Compute

The script downloads the model to the rented machine. How to run it is in the model card.

  1. 1. Rent a machine with a fitting GPU

    Open Massed Compute
  2. 2. Paste the script into the machine's terminal

    Connect over SSH (the Massed Compute panel shows the command) or open a terminal on the machine's desktop.

    HF_TOKEN / CIVITAI_TOKENThe script starts with token lines: paste yours between the quotes. The Hugging Face token is for gated models, the Civitai one for Civitai models.

    # tisi/spriteforge-v3-comfy
    # ===== Tokens: paste yours between the quotes =====
    export HF_TOKEN=""        # huggingface.co/settings/tokens, for gated models
    # ==========================================
    mkdir -p ~/models
    curl -LsSf https://astral.sh/uv/install.sh | sh
    source "$HOME/.local/bin/env"
    uv tool install "huggingface_hub[cli]"
    hf download Qwen/Qwen-Image-Edit-2511 --local-dir ~/models/Qwen-Image-Edit-2511
    hf download tisi/spriteforge-v3-comfy --local-dir ~/models/spriteforge-v3-comfy
    # The model is in ~/models. How to run it is in the model card (link above).
    Add more LoRAs to Qwen-Image-Edit-2511
  3. 3. Use it

    The model is in ~/models. How to run it is in the model card (link above).

Stop the machine in the panel when you are done: it bills by the minute while it runs.

How to speed it up

Picked for this model type. Choose your hardware above to see what fits you exactly.

  • FlashAttention 44.0.0b40from November 18, 2026new release

    fastest exact attention on B200

    Kernels for Blackwell (beta). Exact attention without quantisation — fine for everything, LLMs included.

    How to enable: pip install flash-attn-4; vLLM and SGLang pick it up themselves.

    Earlier generations (2)
    • FlashAttention 3 — The Hopper (H100/H200) version with FP8.
    • FlashAttention 2 — The standard for RTX 30/40/50 and A100: exact attention without the memory cost of long contexts.
  • torch.compile2.15.0from October 28, 2026new release

    10–30% for free

    Compiles the model for your GPU. Slower first run, faster afterwards.

    How to enable: model = torch.compile(model); in ComfyUI the TorchCompileModel node.

Model family

Requires the base model

Qwen/Qwen-Image-Edit-2511

This model is a LoRA / adapter. Download the base model first, then load this one on top.

New posts about versions of Qwen-Image-Edit-2511 (quants, finetunes, LoRAs) will show up under the bell.

Quantised versions 12All on Hugging Face
Fine-tunes 4All on Hugging Face
LoRAs and adapters 12All on Hugging Face

About the model

tisi/spriteforge-v3-comfy
Author
tisi
Category
Other
Released
Posted

On NeyroLearn

Hugging Face

00

Ask AI about this model

Answered by Gemini 4 · the chat is kept only in this tab

Comments