torchao INT8/FP8 weight-only quantized MiniMax-H3 (video+audio omni) for consumer GPUs. Group-offload recipe; load with diffusers (main) + torchao.
Chohan
abhishekchohan
AI & ML interests
None yet
Recent Activity
updated a model about 13 hours ago
abhishekchohan/Muse-Glimmer-30B-GPTQ-INT4 published a model about 13 hours ago
abhishekchohan/Muse-Glimmer-30B-GPTQ-INT4 updated a collection 3 days ago
MiniMax-H3 — INT8 & FP8 Quantizations (torchao)Organizations
None yet
models 89
abhishekchohan/Muse-Glimmer-30B-GPTQ-INT4
Image-Text-to-Text • 9B • Updated
abhishekchohan/minimax-h3-fp8
Text-to-Video • Updated • 23
abhishekchohan/minimax-h3-int8
Text-to-Video • Updated • 37 • 1
abhishekchohan/flux2-klein-4b-fp8
Text-to-Image • Updated • 29
abhishekchohan/flux2-klein-9b-fp8
Text-to-Image • Updated • 33
abhishekchohan/maesar-9B-v2
9B • Updated • 147
abhishekchohan/gemma-4-E4B-it-W4A16
9B • Updated • 139k
abhishekchohan/Qwen3.6-35B-A3B-Abliterated-AWQ
36B • Updated • 4.23k • 2
abhishekchohan/Qwen3.6-27B-abliterated-W4A16
28B • Updated • 11
abhishekchohan/Qwen3.6-35B-A3B-Abliterated-W4A16
35B • Updated • 36 • 1
datasets 0
None public yet