Running a 20B Video Diffusion Transformer on Consumer Hardware: A Deployment and Measurement Report for MiniMax H3 on Apple Silicon

We report the end-to-end deployment and measurement of MiniMax H3, a 20.38-billion-parameter video diffusion transformer (DiT), running entirely on consumer Apple Silicon hardware: an Apple Mac mini with M5 Pro chip and 64 GiB of unified memory. Using the official bf16-to-Q8 (8-bit, group size 64) quantization path executed locally on the machine, we produced a working text-to-video pipeline with no cloud dependency. A 15.08-second 576x1024 clip rendered in 4838.49 seconds with a measured MLX peak of 27.467 GiB. We further operated the system as an unattended production pipeline, completing 17 clips in 26.49 hours with zero render failures. A blind human review of the 17 production clips averaged 3.12, 3.18, and 2.94 out of 5 for structural integrity, temporal consistency, and visible artifacts.

Authors

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-10-19
DOI
https://doi.org/10.5281/zenodo.23484925
Primary Topic
Generative Adversarial Networks and Image Synthesis
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Running a 20B Video Diffusion Transformer on Consumer Hardware: A Deployment and Measurement Report for MiniMax H3 on Apple Silicon

Luke Shi, Steven Shi
Zenodo (CERN European Organization for Nuclear Research)
Generative Adversarial Networks and Image Synthesis
article

Running a 20B Video Diffusion Transformer on Consumer Hardware: A Deployment and Measurement Report for MiniMax H3 on Apple Silicon

Luke Shi, Steven Shi
article en

Abstract

We report the end-to-end deployment and measurement of MiniMax H3, a 20.38-billion-parameter video diffusion transformer (DiT), running entirely on consumer Apple Silicon hardware: an Apple Mac mini with M5 Pro chip and 64 GiB of unified memory. Using the official bf16-to-Q8 (8-bit, group size 64) quantization path executed locally on the machine, we produced a working text-to-video pipeline with no cloud dependency. A 15.08-second 576x1024 clip rendered in 4838.49 seconds with a measured MLX peak of 27.467 GiB. We further operated the system as an unattended production pipeline, completing 17 clips in 26.49 hours with zero render failures. A blind human review of the 17 production clips averaged 3.12, 3.18, and 2.94 out of 5 for structural integrity, temporal consistency, and visible artifacts.

Zenodo (CERN European Organization for Nuclear Research)
Openalex Percentile: Top 15%
Generative Adversarial Networks and Image Synthesis
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.