Appgithub-com-larryvrh-comfyui-minimax-h3-turbo

Install Pinokio

Log in Register

Log in Register

ComfyUI-MiniMax-H3-Turbo

https://github.com/larryvrh/comfyui-minimax-h3-turboupdated 8/7/2026, 11:30:12 PMindexed 8/12/2026, 3:29:54 PM

Contribute to Larryvrh/ComfyUI-MiniMax-H3-Turbo development by creating an account on GitHub.

Loading community details…

Community

Sort by

Post about ComfyUI-MiniMax-H3-Turbo...Post

Loading...

Loading project activity…

Pinokio

Pinokio Apps Using This Repo

No Pinokio apps using this repo yet.

Community tagsNone yet.

Check-ins (0)

Platforms (0)

No reports yet.

Arch (0)

No reports yet.

GPU (0)

No reports yet.

RAM (0)

No reports yet.

VRAM (0)

No reports yet.

Recent commits

Merge pull request #16 from afkaf/main

LarryVRHlast month4274783

Merge pull request #10 from bkunbargi/fix-audio-ref-adaln-rows

LarryVRHlast month546b502

README: v4-600 vs v1-850 decision tree — v1-850 is friendlier for 4-step heavy motion

developerlast month55fee86

README: overhaul — v4-600 is the current recommended checkpoint; clarify 4-8 step range

developerlast month2dcbf1b

v1.2.3: auto-adapt the sampler to ComfyUI's native ModelSamplingAV; fix skipped int8 fc2 LoRA

developerlast month77bfa67

v1.2.2: fix pruned-base unload crash + bypass VRAM regression (issue #4); add low_vram switch

developerlast montha7624b4

README: LoRA-strength tip for the sharpness/artifact trade-off (v1.2.1)

developerlast month55f85c6

Apply LoRA at run time (bypass) instead of merging; device-safe adaln; restore it/s progress bar; add forward observability (v1.2.0)

developerlast monthcc6eb25

README: ask users to keep the node updated

Larryvrhlast month96cc1dd

Pruned/curve base support: run-time adaln injection, one LoRA covers all bases (v1.1.0)

Larryvrhlast monthe7ad532