Text-to-Image
Diffusers
Safetensors
MiniMax H3
VideoX Fun
controlnet
controlnet-union
image-to-image
ml-intern
Instructions to use akhaliq/MiniMax-H3-Image-Fun-Controlnet-Union-2.0 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use akhaliq/MiniMax-H3-Image-Fun-Controlnet-Union-2.0 with Diffusers:
pip install -U diffusers transformers accelerate
from diffusers import ControlNetModel, StableDiffusionControlNetPipeline controlnet = ControlNetModel.from_pretrained("akhaliq/MiniMax-H3-Image-Fun-Controlnet-Union-2.0") pipe = StableDiffusionControlNetPipeline.from_pretrained( "MiniMaxAI/MiniMax-H3-Image", controlnet=controlnet ) - VideoX Fun
How to use akhaliq/MiniMax-H3-Image-Fun-Controlnet-Union-2.0 with VideoX Fun:
# No code snippets available yet for this library. # To use this model, check the repository files and the library's documentation. # Want to help? PRs adding snippets are welcome at: # https://github.com/huggingface/huggingface.js
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- Draw Things
- DiffusionBee
Commit History
Yesterday
-
Card: skeleton pose-control grid with honest partial verdict
Verified signaturecbd6d3f -
Card: fair-comparison showcase grid (video+ref, ControlNet, image+ref)
Verified signaturece7720a -
Card: 5-col showcase grid (H3-Image column, s060 identity-preserving rows)
Verified signature2d11fa1 -
Card: row-4 scale-0.6 note (identity kept + pose survives)
Verified signature86ae835
2 days ago
-
Card: pose showcase grid (extreme poses -> man in a suit)
Verified signaturefba2054 -
Card: drop vote wording from future-work note
Verified signature1453d53 -
Card: reframe — video model works; image model tried, doesn't steer; fine-tune if demand
Verified signature082c471 -
Card: ComfyUI usage section + 2.0-patch workflow
Verified signature88992dc -
Card: official controls on the H3-Image DiT — prompt-driven, not control-driven (with PSNR table)
Verified signatureb7f40f8 -
Clean card: official demo reproduction as hero, trimmed narrative
Verified signature1ae42e2 -
Model card: all 9 official card demos reproduced + motion-vs-static finding
Verified signature45bf269 -
Model card: full feature coverage — 3 of 8 conditions steer via video stack, inpainting mechanism works, honest matrix with samples
Verified signature757a4c8
3 days ago
-
Model card v7: native pseudo-control recipe — fastest working path, measured
Verified signatureb28f6d9 -
Model card v6: weight-distance proof — P-H3 ≈ video trunk (0.001), ≠ H3-Image (0.4)
Verified signaturee664622 -
Model card v5: ph3-control recipe ported and falsified on base H3-Image
Verified signature18e5f8d -
Model card v4: native DiT verified bit-exact, no steering at any grid; fine-tune recipe proposed
Verified signature337305d -
Model card v3: video-pipeline control WORKS (with on/off evidence); image-DiT port still needs adaptation
Verified signature81a002a -
Model card v2: validation verdict — branch attaches but does not steer; evidence + repro
Verified signature2390667 -
Initial model card: experimental H3-Image + ControlNet-Union 2.0 merge
Verified signature434f968