minimax-h3-masafy-lora — "Masafy"

A character LoRA for MiniMax H3, a ~33B-parameter omni-modal model that generates video and audio jointly. Trained on the Ref2VA variant and applied at inference to its GGUF Q3_K_M build. Trigger token: masafy_character.

The subject is an original red-panda character created by the author. The full technical report (JA/EN), the training dataset and all measured data are on GitHub:

https://github.com/masafykun/minimax-h3-masafy-lora

The finding behind this adapter

Training this adapter failed three times before it succeeded, and the reason was not the one everybody expects. Peak VRAM during training measured 3,959 MiB — 2.8% of the card in use. What actually blocked the run was system RAM: the 41.2 GB model set (20.0 GB transformer + 15.7 GB Qwen3-VL 32B text encoder + 5.5 GB VAEs) is staged in host memory before it reaches the GPU, so hosts with 50 GB of RAM were killed by the OOM reaper while a host with 188 GB completed.

Rent by system RAM, not by VRAM. Advertised RAM tracks neither price nor VRAM.

Checkpoints

File	Steps	Notes
masafy_h3_ref2va_lora_v1_000000300.safetensors	300	style already largely acquired
masafy_h3_ref2va_lora_v1_000000400.safetensors	400
masafy_h3_ref2va_lora_v1_000000500.safetensors	500	recommended
masafy_h3_ref2va_lora_v1_000000600.safetensors	600
masafy_h3_ref2va_lora_v1_000000700.safetensors	700	mild degradation
masafy_h3_ref2va_lora_v1.safetensors	800	final; no better than 500

Recommended: checkpoint 500 at strength 0.8. Cross-validation over strength (0.0–1.2) and training length (300–800) put the optimum there. Strength trades style fidelity against structural accuracy: raising it moves the style toward the target while degrading hands, and at 1.2 a heavy dark contour appears around the character — the die-cut border of the sticker artwork it was trained on.

800 steps at strength 1.0, the setting used initially, was excessive in both.

Usage (ComfyUI)

Place a checkpoint in ComfyUI/models/loras/ and insert LoraLoaderModelOnly after UnetLoaderGGUF:

UnetLoaderGGUF → LoraLoaderModelOnly → BasicGuider
                                     → BasicScheduler

Re-point both BasicGuider and BasicScheduler. Patching only one leaves the guider and the scheduler looking at different models, and it fails quietly.

Prompt with the trigger followed by the appearance:

masafy_character, a cute anthropomorphic red panda wearing brown aviator goggles
and a pink patterned scarf, waving at the camera with a big smile, standing in a
sunny park, smooth character animation, clean bright colors

The trigger alone is not sufficient. With 12 training images all captioned with the same appearance clause, the caption text carries real weight; omitting the appearance description does not produce a usable result.

What the adapter contributes

Under reference-image (R2V) conditioning the reference already pins the character: without the adapter the goggles, the scarf and the pose all render correctly and without flicker. What changes is the style — flat cel illustration with the adapter, photoreal fur without it. The return on this adapter is rendering the character in the author's own style, not rendering the character at all.

The adapter also reshapes backgrounds, because captions described whole images.

Requirements

Inference VRAM	9.3 GB (fits a 12 GB consumer GPU)
Generation time	~13 min per 5 s clip at 864×480 on an RTX 3060
Base model	MiniMax H3 Ref2VA — not redistributed here
Quantization	trained on pruned INT8, applies cleanly to GGUF Q3_K_M (208/208 modules, zero key mismatches)

License and attribution

The adapter is a derivative work of MiniMax H3 and its use is subject to the MiniMax H3 Community License Agreement (text). Commercial use is permitted and requires attribution to "MiniMax H3". The license carries a territorial clause; consult the official terms for your jurisdiction.

Weights were obtained from Comfy-Org/MiniMax-H3, the ComfyUI redistribution of MiniMaxAI/MiniMax-H3. Neither is redistributed here.

Training additionally required the official training adapter, ostris/minimax_h3_training_adapter, and the ai-toolkit trainer — thanks to ostris for both.

The character "Masafy" is an original character created by the author. The design remains the author's work — please credit the author if you use it.

Citation

@techreport{suzuki2026h3lora,
  author = {Suzuki, Masato},
  title  = {Character LoRA Fine-tuning of a 33B Omni-modal Video Model:
            System RAM, Not VRAM, Was the Binding Constraint},
  year   = {2026},
  url    = {https://github.com/masafykun/minimax-h3-masafy-lora}
}