Unverified50% confidenceFactExact time
Locally deploying MiniMax H3 requires downloading five model files: the main model h3f12t8, the all-purpose reference model h3aef2vnt8 (both choosing the int8 version), the Qianwen minimax-h3vfp4 text encoder, and two VAE decoders for video and audio.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/16/2026
First Seen
Valid until: 1/14/2027
Sources
MiniMax H3 Local Deployment Tutorial: Building an AI Video Workstation with ComfyUI
bilibiliAgent视频教程10/16/2026
Related Entities
Related Claims
UnverifiedMiniMax released the H3 open-source video generation model, which can run with as little as 8 GB of VRAM74% similarUnverifiedThe tail-frame generation module of MiniMax H3 is 7.6 GB in file size; it is not the complete model but is a required component.66% similarUnverifiedRunning the MiniMax H3 base model is recommended with a GPU with 24GB or more VRAM, such as an NVIDIA RTX 4090 or A100.66% similarUnverifiedMiniMax H3's weights total 67 GB, comprising a 34 GB DiT backbone, a 27 GB Qwen3-VL encoder, plus VAE65% similarUnverifiedThe MiniMax H3 community version is already capable of running on GPUs with 8GB or 12GB of VRAM.65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1044942API
curl https://kongchang.com/api/v1/knowledge/claims/1044942MCP
get_claim(id=1044942)