Image-Text-to-Video
Diffusers
Safetensors
MiniMaxH3ModularPipeline
text-to-video
image-to-video
video-to-video
text-to-audio-video
image-to-audio-video
image-text-to-audio-video
video-to-audio-video
audio-to-audio-video
audio-video-generation
multimodal
synchronized-audio-video
reference-to-audio-video
Will the MiniMax-H3 Post-Training Pipeline Be Open-Sourced?
#51
by The000FOOL000 - opened
Hi MiniMax team, thank you for releasing the model weights!
Are there any plans to open-source the post-training resources for MiniMax-H3 as well?
- Post-training / RL training code and recipes
- SFT/RL datasets or data-generation pipelines
- Reward functions or verifiers
- Training configs and hyperparameters
- Any other resources needed to reproduce the post-training process
Even a partial release or an indication of whether you plan to release these in the future would be greatly appreciated.
Thank you!
Not decided
ryanlee-dev changed discussion status to closed