Watch the tutorial video on how to build an anime-style short using this workflow:<div data-youtube-video><iframe width="640" height="480" allowfullscreen="true" autoplay="false" disablekbcontrols="false" enableiframeapi="false" endtime="0" ivloadpolicy="0" loop="false" modestbranding="false" origin playlist src="https://www.youtube.com/embed/mmuFwMeKUB0?rel=1" start="0"></iframe></div>A no-nonsense high level T2V / I2V / FFLF / REF2V workflow for Minimax H3 with lots of options &amp; togglable quality of life features.<ul><li>Toggle between the [T2V / I2V / FFLF] model &amp; the [REF2VA] model easily</li><li>Added Forced Custom Audio: Lipsyncing to custom dialogue/music is now easy for T2V/I2V/REF2VA!</li><li>Temporal Upsampler 2nd Pass Option! [<a target="_blank" rel="ugc" href="https://github.com/matlowai/ComfyUI-MAINodes">https://github.com/matlowai/ComfyUI-MAINodes</a>]</li><li>New Image Loader allows cropping/setting max megapixels directly inside node: <a target="_blank" rel="ugc" href="https://github.com/obvpm/comfyui-obvpm">https://github.com/obvpm/comfyui-obvpm </a>(Must install via unzipping to custom_nodes folder or through comfy-manager's Install via Git option, it's too new to be on the comfy-manager's index)</li><li>Kijai's Preview Override (put <a target="_blank" rel="ugc" href="https://huggingface.co/Kijai/MiniMax-H3-TAE/blob/main/vae_approx/taeh3.safetensors">https://huggingface.co/Kijai/MiniMax-H3-TAE/blob/main/vae_approx/taeh3.safetensors</a> in /vae_approx/ and set as the custom vae for non-pixelated previews)</li><li>Speedup: Sage-Attn + New Sol-Attn [<a target="_blank" rel="ugc" href="https://github.com/kijai/ComfyUI-SolAttn_triton">https://github.com/kijai/ComfyUI-SolAttn_triton</a>]</li><li>Speedup node: EasyCache / Spectrum [<a target="_blank" rel="ugc" href="https://github.com/xmarre/ComfyUI-Spectrum-MiniMax-H3">https://github.com/xmarre/ComfyUI-Spectrum-MiniMax-H3</a>]</li><li>Turbo LoRA (t2v/i2v but works with ref2va if you don't mind [some] audio degradation)<ul><li>New (8/11/26) Lightx2v 1.0 release [<a target="_blank" rel="ugc" href="https://huggingface.co/lightx2v/Minimax-h3-Turbo/tree/main">https://huggingface.co/lightx2v/Minimax-h3-Turbo/tree/main</a>]</li></ul></li><li>Film VFI Frame Interpolation (24 fps -&gt; 48 fps)</li><li>Easily togglable reference fields: 4 pictures, 2 audio, 1 video</li></ul>Use this workflow if you:<ul><li>Are an AI Filmmaker who values character/scene consistency</li><li>Want to be able to use FFLF effectively, seamlessly extend videos, or use character reference sheets.</li><li>Want to edit videos, or copy &amp; use motion from reference videos</li><li>Want fine-tuned control over your shots visuals and audio.</li></ul>If you wish to generate simple one-off video clips like Will Smith eating spaghetti, the t2v/i2v model that you can toggle to in this workflow will do that for you.Kijai's int8 convrot video vae: <a target="_blank" rel="ugc" href="https://huggingface.co/Kijai/MiniMax-H3-experimental/blob/main/minimax_h3_video_vae_int8_convrot.safetensors">https://huggingface.co/Kijai/MiniMax-H3-experimental/blob/main/minimax_h3_video_vae_int8_convrot.safetensors</a>UPDATE YOUR COMFY CUDA VERSION TO 13.0. If you start comfy and see cu130:<pre><code>[INFO] pytorch version: 2.13.0+cu130</code></pre>Then you are good to go! But if you are using cu126, then ALL your gens with the best version of Minimax's model (INT8 convrot) will be 2x slower than they should be due to inefficient comfy-kitchen operations! Keep in mind if you update your cuda, you will need to reinstall Sage Attention! Download the correct wheel from here <a target="_blank" rel="ugc" href="https://wildminder.github.io/AI-windows-whl/">https://wildminder.github.io/AI-windows-whl/</a>ALL MODEL / NODE LINKS ARE IN THE WORKFLOW NOTES OR EASILY INSTALLABLE THROUGH COMFY-UI MANAGER.If you appreciate what I'm doing, please consider following/subscribing on my patreon (free) which gets access to all my work early.<a target="_blank" rel="ugc" href="https://www.patreon.com/cw/foxfuressence">https://www.patreon.com/cw/foxfuressence</a>Also my youtube where I make AI filmmaking tutorials plz &amp; thx:<a target="_blank" rel="ugc" href="https://www.youtube.com/@foxfuressence">https://www.youtube.com/@foxfuressence</a>

Minimax H3 T2V / I2V / REF2V Advanced Filmmaking Workflow | All Speedups + QoL Features

Edit: 8/16/26: Fixed bug where Audio 1 wasn't plugged in to H3 REF2VA node. <ul><li>Adds Comfy-Kitchen Attention Speed Up</li><li>Adds Temporal Upsampling De-Rope (fixes fast motion blur/smudge) [<a target="_blank" rel="ugc" href="https://github.com/matlowai/ComfyUI-MAINodes">https://github.com/matlowai/ComfyUI-MAINodes</a>]</li><li>Replaces RIFE with built in comfy node Film VFI</li><li>Adds Model Shift node</li></ul>This is a pretty large update. The Temporal Upsampling 2nd Pass option improves action scenes drastically.

Files with SHA256: 58ef7b13cf007e95c0445cbfc745e58749b64a67001e583796efc2fce998506f