Popular repositories Loading
-
minimax-h3-chunk-star7
minimax-h3-chunk-star7 PublicVRAM-efficient MiniMax H3 for ComfyUI: QKV/RoPE/MLP activation chunking, automatic OOM chunk reduction, and CK/SLA/Sol/Hybrid attention for SM75 and SM80+ GPUs.
-
minimax-h3-fp16-exact-star7
minimax-h3-fp16-exact-star7 PublicA ComfyUI node providing an FP16 compatibility path for MiniMax H3 on NVIDIA Turing GPUs such as the RTX 2080 Ti.
-
star7-director
star7-director PublicStar7 Director: a ComfyUI canvas director desk for MiniMax H3, storyboard queues, materials and staged generation.
Python 5
-
minimax-h3-enhanced-loader-star7
minimax-h3-enhanced-loader-star7 PublicArchitecture-aware ComfyUI loader for official, third-party, quantized, and FastH3 MiniMax H3 models with FP16 protection and VSA support.
Python 3
If the problem persists, check the GitHub status page or contact support.

