ComfyUI MiniMax H3 Text-to-Video, Image-to-Video, and Reference-to-Video Workflows

This page covers the three base MiniMax H3 templates plus advanced native-node techniques. For capabilities, tuning, and speedups, see the MiniMax H3 overview.

MiniMax H3 Text to Video (T2V)

Generate videos from text prompts with native stereo audio.

Run on Comfy Cloud

Open in Comfy Cloud

Download Workflow

Download JSON or search “MiniMax H3 T2V” in Template Library

Model downloads

Diffusion Model: minimax_h3_fl2va_pruned_int8_convrot

Place in ComfyUI/models/diffusion_models/

Text Encoder: qwen3vl_32b_minimax_h3_nvfp4_awq

Place in ComfyUI/models/text_encoders/

VAE: minimax_h3_video_vae_int8_convrot

Place in ComfyUI/models/vae/

VAE: minimax_h3_audio_vae_fp32

Place in ComfyUI/models/vae/

LoRA: minimax_h3_fl2v_turbo_8step_v1.0_comfyui_bf16

Required by the workflow’s model scan; powers the optional Lightning LoRA turbo mode. Place in ComfyUI/models/loras/

Model storage

ComfyUI/
├── 📂 models/
│   ├── 📂 diffusion_models/
│   │   └── minimax_h3_fl2va_pruned_int8_convrot.safetensors
│   ├── 📂 text_encoders/
│   │   └── qwen3vl_32b_minimax_h3_nvfp4_awq.safetensors
│   ├── 📂 vae/
│   │   ├── minimax_h3_video_vae_int8_convrot.safetensors
│   │   └── minimax_h3_audio_vae_fp32.safetensors
│   ├── 📂 loras/
│   │   └── minimax_h3_fl2v_turbo_8step_v1.0_comfyui_bf16.safetensors
│   └── 📂 embeddings/
│       └── minimaxh3_art_is_explosion.safetensors

Prompting tips

Scene first: Overall scene, then timed shots
One block: Shots, camera, dialogue, SFX, and music together
Resolution: Native 16:9 is 1344×768 (768px short edge); round to multiples of 32
Duration: Snaps to 17k+5 frames at 24fps
First/last frame: Wire images to first_frame / last_frame on MiniMaxH3ImageToVideo for FL2VA
Turbo: Default 20 steps; enable turbo_mode for 8-step Lightning LoRA (faster, slightly softer motion/audio). Raise steps (e.g. 25) for motion quality

Official base guide: prompt guide.

MiniMax H3 Image to Video (I2V)

Generate videos from an input image, with optional first/last-frame keyframes.

Run on Comfy Cloud

Open in Comfy Cloud

Download Workflow

Download JSON or search “MiniMax H3 I2V” in Template Library

Input Image: transparent_rgb_gaming_mouse.png

Download the default input image, or use your own image.

Model downloads

Diffusion Model: minimax_h3_fl2va_pruned_int8_convrot

Place in ComfyUI/models/diffusion_models/

Text Encoder: qwen3vl_32b_minimax_h3_nvfp4_awq

Place in ComfyUI/models/text_encoders/

VAE: minimax_h3_video_vae_int8_convrot

Place in ComfyUI/models/vae/

VAE: minimax_h3_audio_vae_fp32

Place in ComfyUI/models/vae/

Turbo LoRA: minimax_h3_fl2v_turbo_8step_v1.0_comfyui_bf16

Place in ComfyUI/models/loras/. Used when turbo_mode is enabled.

Model storage

ComfyUI/
├── 📂 models/
│   ├── 📂 diffusion_models/
│   │   └── minimax_h3_fl2va_pruned_int8_convrot.safetensors
│   ├── 📂 text_encoders/
│   │   └── qwen3vl_32b_minimax_h3_nvfp4_awq.safetensors
│   ├── 📂 loras/
│   │   └── minimax_h3_fl2v_turbo_8step_v1.0_comfyui_bf16.safetensors
│   ├── 📂 vae/
│   │   ├── minimax_h3_video_vae_int8_convrot.safetensors
│   │   └── minimax_h3_audio_vae_fp32.safetensors
│   └── 📂 embeddings/
│       └── minimaxh3_art_is_explosion.safetensors

Prompting tips

Keyframes: Optional first_frame / last_frame; motion is generated between them
Prompt: Shots, motion, and audio in one block
Resolution / duration: the same rules as T2V (1344×768 native target, 17k+5 grid)
Turbo: turbo_mode uses turbo_steps (default 8) and turbo_model_strength (default 1.0) instead of 20 steps

first_frame is resized to the canvas without keeping its aspect ratio, so an off-aspect image comes out stretched. last_frame is center-cropped instead. The I2V template opens at 1:1 and 0.4 MP (640×640), so set the aspect ratio to your image and raise Megapixels for a final render. Official base guide: prompt guide.

MiniMax H3 Reference to Video (R2V)

Lock character, style, motion, camera, or voice from reference images, videos, and audio.

Run on Comfy Cloud

Open in Comfy Cloud

Download Workflow

Download JSON or search “MiniMax H3 R2V” in Template Library

Reference Image: red_superboy_on_city_roof.png

Character reference for the workflow, or use your own images.

Reference Image: mecha_dragon_lightning.png

Style and subject reference for the workflow, or use your own images.

Model downloads

Diffusion Model: minimax_h3_ref2va_pruned_int8_convrot

Place in ComfyUI/models/diffusion_models/

Text Encoder: qwen3vl_32b_minimax_h3_nvfp4_awq

Place in ComfyUI/models/text_encoders/

VAE: minimax_h3_video_vae_int8_convrot

Place in ComfyUI/models/vae/

VAE: minimax_h3_audio_vae_fp32

Place in ComfyUI/models/vae/

LoRA: minimax_h3_ref2v_turbo_4step_v0.1_comfyui_bf16

Required by the workflow’s model scan; powers the optional Lightning LoRA turbo mode. Place in ComfyUI/models/loras/

Model storage

ComfyUI/
├── 📂 models/
│   ├── 📂 diffusion_models/
│   │   └── minimax_h3_ref2va_pruned_int8_convrot.safetensors
│   ├── 📂 text_encoders/
│   │   └── qwen3vl_32b_minimax_h3_nvfp4_awq.safetensors
│   ├── 📂 vae/
│   │   ├── minimax_h3_video_vae_int8_convrot.safetensors
│   │   └── minimax_h3_audio_vae_fp32.safetensors
│   ├── 📂 loras/
│   │   └── minimax_h3_ref2v_turbo_4step_v0.1_comfyui_bf16.safetensors
│   └── 📂 embeddings/
│       └── minimaxh3_art_is_explosion.safetensors

Prompting tips

Tags: Reference inputs as <Picture 1>, <Video 1>, <Audio 1> (ref_image_0 → <Picture 1>). Node order: images, then videos, then audio
Assign roles: Say which reference drives identity, style, motion, camera, or voice
Limits: Up to 9 images, 3 videos, 3 standalone audio clips
Video FPS: Reference video is read as 24fps with no metadata. Convert other frame rates to 24fps before load
Video sound: Picture and soundtrack are separate references. Wire ref_video_audio_N to match ref_video_N or the file’s audio is not used as <Audio N>
Audio numbering: Video soundtracks get <Audio N> first (video order), then standalone clips. Video and picture indices are independent
VAE wiring: Video soundtrack needs vae + audio_vae; standalone audio needs audio_vae only
Audio roles: e.g. use <Audio 1> exactly as it is vs follow the timbre of <Audio 2>
Voice: Audio carries timbre, not words. Specify delivery; one audio ref per speaker when assignment must be exact
ref_image_size: Default match (fit to canvas). max keeps 2048px short edge and is much slower; it does not upscale small files (round sides to 32). Pre-scale to 2048 short edge when needed
Weights: R2V uses ref2va, not the T2V/I2V fl2va model
Turbo: Optional 4-step Lightning LoRA (faster, weaker reference/audio). Prefer 20–25 steps when references must hold
Style LoRAs: Compete with references; use low strength (~0.5) on base schedule when adherence matters

Official ref guide: prompt guide.

Advanced workflows with the native nodes

Requires a recent ComfyUI build for the features below.

Anchoring guides at any frame (#15439)

MiniMaxH3AddGuide anchors a guide at any frame_idx (negative = from end), not only first/last frame. Wire positive and latent from an H3 node into MiniMaxH3AddGuide. Inputs:

image: Still or clip (multi-frame batches crop to valid 17k+5 lengths; <5 frames uses first image only)
audio: Soundtrack at the same frame index, cropped to remaining duration
frame_idx: Anchor position

Connect vae for image guides and audio_vae for audio. Chain nodes for multiple anchors. Example: first 22 frames of a clip at frame 0 continues both streams; a still at frame 60 forces the shot through that pose. This is conditioning, not video-to-video: sampling still starts from the workflow empty latent. denoise does not scale guide strength like V2V. To restyle a whole clip, use R2V; to edit regions, use Fun ControlNet Union masks. See the Multiframe Reference template.

Inpainting and extension with latent noise masks (#15375)

Per-token noise masks on video and audio latents let you regenerate part of a clip. Connect a mask to the sampler denoise_mask: 0 = keep, 1 = regenerate. Video masks snap to the 2×2 latent patch grid; audio masks to whole latent frames. Use for inpainting, removal, or extension while holding the rest stable.

​MiniMax H3 Text to Video (T2V)

Run on Comfy Cloud

Download Workflow

​Model downloads

Diffusion Model: minimax_h3_fl2va_pruned_int8_convrot

Text Encoder: qwen3vl_32b_minimax_h3_nvfp4_awq

VAE: minimax_h3_video_vae_int8_convrot

VAE: minimax_h3_audio_vae_fp32

LoRA: minimax_h3_fl2v_turbo_8step_v1.0_comfyui_bf16

​Model storage

​Prompting tips

​MiniMax H3 Image to Video (I2V)

Run on Comfy Cloud

Download Workflow

Input Image: transparent_rgb_gaming_mouse.png

​Model downloads

Diffusion Model: minimax_h3_fl2va_pruned_int8_convrot

Text Encoder: qwen3vl_32b_minimax_h3_nvfp4_awq

VAE: minimax_h3_video_vae_int8_convrot

VAE: minimax_h3_audio_vae_fp32

Turbo LoRA: minimax_h3_fl2v_turbo_8step_v1.0_comfyui_bf16

​Model storage

​Prompting tips

​MiniMax H3 Reference to Video (R2V)

Run on Comfy Cloud

Download Workflow

Reference Image: red_superboy_on_city_roof.png

Reference Image: mecha_dragon_lightning.png

​Model downloads

Diffusion Model: minimax_h3_ref2va_pruned_int8_convrot

Text Encoder: qwen3vl_32b_minimax_h3_nvfp4_awq

VAE: minimax_h3_video_vae_int8_convrot

VAE: minimax_h3_audio_vae_fp32

LoRA: minimax_h3_ref2v_turbo_4step_v0.1_comfyui_bf16

​Model storage

​Prompting tips

​Advanced workflows with the native nodes

​Anchoring guides at any frame (#15439)

​Inpainting and extension with latent noise masks (#15375)

MiniMax H3 Text to Video (T2V)

Model downloads

Model storage

Prompting tips

MiniMax H3 Image to Video (I2V)

Model downloads

Model storage

Prompting tips

MiniMax H3 Reference to Video (R2V)

Model downloads

Model storage

Prompting tips

Advanced workflows with the native nodes

Anchoring guides at any frame (#15439)

Inpainting and extension with latent noise masks (#15375)