Wuli's Gemma-4 Prompt Extender for MiniMax H3 Runs in ComfyUI

ComfyUI Wikinews

Wuli-Art fine-tuned Gemma-4-12B into a MiniMax H3 prompt extender: one weight set for FL2VA and Ref2VA, fp8 at 13 GB, with two ComfyUI workflows.

Wuli-Art released Gemma4-for-MiniMax-H3-Prompt-Extend on 2026-11-16: a Gemma-4-12B fine-tune that rewrites a plain idea into the audiovisual prompt MiniMax H3 expects. It ships bf16, fp8 and int8 weights plus two ComfyUI workflows, and those graphs use core nodes only.

What it does

An H3 prompt is one block that has to carry the shot, the action over time, the camera move and the soundscape. Writing that by hand is the slow part. This model writes it from your sentence plus whatever frames or references you already loaded.

One weight set serves both input formats, so you do not need a different extender for first/last-frame work and for reference work.

H3 input formatCoversWhat the extender adds
FL2VAT2VA, I2VA, L2VA, FL2VAscene, action, camera movement, sound, frame constraints
Ref2VAreference-conditioned generationsubject definitions, reference roles, what to keep

How to run it

  1. Put gemma4_12b_h3_pe_fp8_scaled.safetensors (13.0 GB) or the int8 file in ComfyUI/models/text_encoders/. The bf16 file is 23.9 GB.
  2. Drag the matching workflow onto the canvas: FL2VA or Ref2VA. They are alternatives, not two stages.
  3. Point the UNET, text encoder, VAE and LoRA loaders at your own H3 files, write the prompt, run. The output is text, which you feed to your normal H3 graph.
The enhanced prompt comes back in English even when you write the request in Chinese. The model card lists zh and en as its languages.

Version requirement: a ComfyUI build with TextGenerateV2 and Gemma-4 text-encoder support. v0.46.0 has both.

Downloads

FileSizeWhere it goes
gemma4_12b_h3_pe_fp8_scaled.safetensors13.0 GBmodels/text_encoders/
gemma4_12b_h3_pe_int8.safetensors13.0 GBmodels/text_encoders/
gemma4_12b_h3_pe_bf16.safetensors23.9 GBmodels/text_encoders/

Compared with the other prompt writers

The model card's demo runs five settings on the same inputs: no extension, plain Gemma-4-12B, MiniMax's official PE, lightx2v's rewriter, and this model. Case 04 starts from one 31-character Chinese line about two cranes fighting, plus a first frame.

No prompt extensionMiniMax official PEGemma-4-12B extender
No extensionMiniMax official PEGemma-4-12B extender

Length is the visible difference. On this case the plain Gemma-4 baseline wrote 1,097 characters, lightx2v's rewriter 1,180, this model 2,102, and the official PE 4,296.

A 5-second 1024×768 sample generated from this extender's prompt.

Availability

Comments

Sign in with GitHub to join the discussion.

Loading comments…
Wuli's Gemma-4 Prompt Extender for MiniMax H3 Runs in ComfyUI | ComfyUI Wiki