Michael Becker mdbecker

Detailed Guide for Running Hunyuan Video T2V on Your M4 MacBook

This guide shows you how to run the Hunyuan Video text‑to‑video (T2V) workflow on your M4 MacBook (128 GB RAM with a top‑end GPU). We’ll use the pre‑built dmg file to install ComfyUI (avoiding the need to compile C++ code), set up the workflow in “fast video” mode (reducing inference steps from ~20 to 6–8), and cover downloading and organizing the required model files.

1. Overview

You will install ComfyUI using its pre‑built dmg installer, then configure the Hunyuan Video workflow from the available collection. We focus on using T2V (text‑to‑video) with the “fast video” mode to achieve faster iteration times. With your M4 MacBook’s ample resources, you can experiment confidently while monitoring system performance.

Keybase proof

I hereby claim:

I am mdbecker on github.
I am mdbecker (https://keybase.io/mdbecker) on keybase.
I have a public key whose fingerprint is 3269 BEE3 B3B2 23ED 1478 2F5B 73DE 3334 40FF FBF7

To claim this, I am signing this object:

	import hashlib
	import json
	import random
	from datetime import datetime

	import diffusers
	import piexif
	import torch
	from diffusers import FluxPipeline
	from PIL import Image

	import mailbox
	import csv
	import email.utils
	from collections import defaultdict
	import argparse
	from pathlib import Path
	from bs4 import BeautifulSoup
	from tqdm import tqdm

	# Constants for field names

	"""
	Fixes https://github.com/scikit-learn/scikit-learn/issues/12052
	CalibratedClassifierGroupCV is a drop in replacment for CalibratedClassifierCV that supports GroupKFold cv.

	This is based off of https://github.com/scikit-learn/scikit-learn/blob/0.24.1/sklearn/calibration.py.
	If you are using a different version of sklearn, you can make similar modifications to your version.

	Example usage:

	```

	initial_susceptible # defaults to 3,600,000 https://github.com/CodeForPhilly/chime/blob/2895a9c4ddcf42b3c96bcf7e03a7e2a15f4983de/src/penn_chime/presentation.py#L200-L206
	initial_infected # https://github.com/CodeForPhilly/chime/blob/2895a9c4ddcf42b3c96bcf7e03a7e2a15f4983de/src/penn_chime/models.py#L25-L27
	initial_recovered # https://github.com/CodeForPhilly/chime/blob/2895a9c4ddcf42b3c96bcf7e03a7e2a15f4983de/src/penn_chime/models.py#L34
	beta = # https://github.com/CodeForPhilly/chime/blob/2895a9c4ddcf42b3c96bcf7e03a7e2a15f4983de/src/penn_chime/models.py#L42-L45
	gamma = # https://github.com/CodeForPhilly/chime/blob/2895a9c4ddcf42b3c96bcf7e03a7e2a15f4983de/src/penn_chime/models.py#L39
	n_days = # User input, default to 60 or something

	def sir(s, i, r, beta, gama, n):
	"""The SIR model, one time step."""
	s_n = (-beta * s * i) + s

	from os import getpid, kill
	from time import sleep
	import re
	import signal

	from notebook.notebookapp import list_running_servers
	from requests import get
	from requests.compat import urljoin
	import ipykernel
	import json

	import seaborn as sns
	import numpy as np
	import matplotlib.pyplot as plt


	class _MiniBoxPlotter(sns.categorical._ViolinPlotter):

	def draw_violins(self, ax):
	"""Draw the violins onto `ax`."""
	for i, group_data in enumerate(self.plot_data):

	from sklearn import metrics

	def binary_cv_metrics(y, preds, m):
	ACC = metrics.accuracy_score(y,preds)
	cm = metrics.confusion_matrix(y,preds)

	m['confusion_matrix'] = cm
	m['Accuracy'] = ACC
	m['F1 score'] = metrics.f1_score(y,preds)
	m['FPR'] = cm[0,1]/(cm[0,:].sum()*1.0)