<h1 id="introducing:-minimax-h3-or-ez-wf">Introducing: MINIMAX H3 | EZ WF </h1>UPDATE: The workflow has been completely redone on 08/08/2026. if you previously downloaded and it didn't work for you, try now. i have tested in a fresh install, it works, if you are having issues you most likely don't have the most up to date version on Comfy from the master branch here: <a target="_blank" rel="ugc" href="https://github.com/Comfy-Org/ComfyUI/branches">https://github.com/Comfy-Org/ComfyUI/branches</a>. do NOT use the releases version 0.31.0, it will not work. Also be sure to update all of the nodes correctly, pull them from GitHub. Check your console for any missing dependencies like sage-attention etc and install them. i will no longer be responding to any issues regarding node pack issues, you just don't have the updated versions or you have something in your setup that conflicts. also make sure your models are compatible, users have been reporting it doesn't work, only to find the are using models that aren't compatible, this model was made with INT8 in mind. If you are having black outputs you need to either remove the start-up flag --use-sage-attention, change to the normal VAE if you are using the INT8 one i linked, or again update comfy to latest version on master branch.<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/89acf9ea-5353-411c-ad55-29fda31069e5/original=true/89acf9ea-5353-411c-ad55-29fda31069e5.jpeg" /> a custom workflow for Minimax's H3 video model. Its not too different to the base one Comfy provides, just has some little tweaks that i like to use, including Sage Attention/Turbo LoRAs and more! This workflow is my personal workflow that I use for all of my Minimax H3 stuff, it may not be to everyone's taste but it works well for me and I'm sure others may find it useful.Generated a 5 second clip with 0.2MP &gt; 1x LTX Upscale at 16:9 in 7:32 on an RTX 3060 12GB/32GB Sys RAM using this WF. base workflow was more than double that time. <img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/ce751169-8628-4d72-bb5e-845b175575d6/original=true/ce751169-8628-4d72-bb5e-845b175575d6.jpeg" />The workflow uses the most optimal boosters there are right now (all optional, you can just run without them)<hr />Works on 12GB VRAM + 32 GB System RAM. not tested on anything below.INCLUDES:<ul><li>3 Ref Images | 2 Ref Audio | 1 Ref Video</li><li>Sage Attention Patch (For Lower VRAM Use)</li><li>Block Cache &amp; Sol-Attn with Optimised Settings</li><li>Bob INT8 Loader (Faster INT8 Loading)</li><li>LTX Refiner</li><li>RTX Upscaler</li><li>ITV/T2V/RTV Modes</li></ul><hr /><h3 id="nodes-needed:">Nodes Needed:</h3>[Bob INT8](<a target="_blank" rel="ugc" href="https://github.com/BobJohnson24/ComfyUI-INT8-Fast">https://github.com/BobJohnson24/ComfyUI-INT8-Fast</a>)[Sol-Att](<a target="_blank" rel="ugc" href="https://github.com/kijai/ComfyUI-SolAttn_triton">https://github.com/kijai/ComfyUI-SolAttn_triton</a>)[KJNodes](<a target="_blank" rel="ugc" href="https://github.com/kijai/ComfyUI-KJNodes">https://github.com/kijai/ComfyUI-KJNodes</a>)[rgthree](<a target="_blank" rel="ugc" href="https://github.com/rgthree/rgthree-comfy">https://github.com/rgthree/rgthree-comfy</a>)[EasyUse](<a target="_blank" rel="ugc" href="https://github.com/yolain/ComfyUI-Easy-Use">https://github.com/yolain/ComfyUI-Easy-Use</a>)[VHS](<a target="_blank" rel="ugc" href="https://github.com/Kosinkadink/ComfyUI-VideoHelperSuite">https://github.com/Kosinkadink/ComfyUI-VideoHelperSuite</a>)[RTX Upscaler](<a target="_blank" rel="ugc" href="https://github.com/Comfy-Org/Nvidia_RTX_Nodes_ComfyUI">https://github.com/Comfy-Org/Nvidia_RTX_Nodes_ComfyUI</a>)[LTXVideo](<a target="_blank" rel="ugc" href="https://github.com/Lightricks/ComfyUI-LTXVideo">https://github.com/Lightricks/ComfyUI-LTXVideo</a>)<h3 id="models-needed:">Models Needed:</h3>-- H3 MODELS --[<a target="_blank" rel="ugc" href="https://huggingface.co/Gluttony10/MiniMax-H3-INT8-CONVROT/blob/main/MiniMax-H3-Ref2VA-int8_convrot.safetensors">REF MODEL</a>][<a target="_blank" rel="ugc" href="https://huggingface.co/Gluttony10/MiniMax-H3-INT8-CONVROT/blob/main/MiniMax-H3-FL2VA-int8_convrot.safetensors">FLF MODEL</a>][<a target="_blank" rel="ugc" href="https://huggingface.co/Gluttony10/MiniMax-H3-INT8-CONVROT/blob/main/qwen3-vl-32b-int8_convrot.safetensors">TE</a>]<a target="_blank" rel="ugc" href="https://huggingface.co/Gluttony10/MiniMax-H3-INT8-CONVROT/blob/main/MiniMax-H3-audio_vae.safetensors">[AUDIO VAE]</a>[<a target="_blank" rel="ugc" href="https://huggingface.co/Kijai/MiniMax-H3-experimental/blob/main/minimax_h3_video_vae_int8_convrot.safetensors">VIDEO VAE</a>]-- LTX MODELS --<a target="_blank" rel="ugc" href="https://civarchive.com/models/2447875/ltx23-10eros?modelVersionId=3109610">[EROSV1.4]</a><a target="_blank" rel="ugc" href="https://huggingface.co/DreamFast/gemma-3-12b-it-heretic-v2/blob/main/comfyui/gemma-3-12b-it-heretic-v2_int8.safetensors">[TE 1]</a> | <a target="_blank" rel="ugc" href="https://huggingface.co/Kijai/LTX2.3_comfy/blob/main/text_encoders/ltx-2.3_text_projection_bf16.safetensors">[TE 2]</a> &lt;-- Need Both<a target="_blank" rel="ugc" href="https://huggingface.co/Kijai/LTX2.3_comfy/blob/main/vae/LTX23_video_vae_bf16.safetensors">[VIDEO VAE]</a><a target="_blank" rel="ugc" href="https://huggingface.co/Kijai/LTX2.3_comfy/blob/main/vae/LTX23_audio_vae_bf16.safetensors">[AUDIO VAE]</a><a target="_blank" rel="ugc" href="https://huggingface.co/Lightricks/LTX-2.3/blob/main/ltx-2.3-spatial-upscaler-x2-1.1.safetensors">[UPSCALER]</a><h3 id="turbo-loras:">TURBO LORAS:</h3>[<a target="_blank" rel="ugc" href="https://huggingface.co/Kijai/MiniMax-H3_comfy/blob/main/loras/minimax_h3_fl2v_lightx2v_turbo_4step_v0.1_comfy.safetensors">LIGHTx</a>][<a target="_blank" rel="ugc" href="https://huggingface.co/drbaph/MiniMax-H3-Turbo-Lora-ComfyUI/blob/main/minimax_h3_turbo_v4_step600_pruned_comfyui.safetensors">LARRY 600</a>]<h3 id="troubleshooting:">Troubleshooting:</h3>I'm missing the LTX Nodes and i get this error, what do i do?<pre><code>ImportError: cannot import name 'pad' from 'kornia.geometry.transform.pyramid' (G:\StabilityMatrix-win-x64\Data\Packages\ComfyUIClean\venv\Lib\site-packages\kornia\geometry\transform\pyramid.py)</code></pre>-- downgrade Kornia in python packages to 0.8.2<hr /><pre><code>The RTX Nodes are missing, even though i have them installed?</code></pre>Go to your python packages and install: <span style="background-color:rgb(11, 20, 22);color:rgb(183, 202, 212);font-family:-apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, &quot;Helvetica Neue&quot;, Arial, &quot;Apple Color Emoji&quot;, &quot;Segoe UI Emoji&quot;, &quot;Segoe UI Symbol&quot;, sans-serif;font-size:14px">nvidia-vfx<hr /><pre><code>I'm missing MiniMaxH3MemoryEfficientSageAttentionPatch</code></pre>Update KJNodes through Github (go to your custom_nodes folder and open a git bash and type: git clone https://github.com/kijai/ComfyUI-KJNodes.gitMAKE SURE YOU ARE UPDATED LIKE SO:<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/ab435bf4-414b-46e3-ac9c-d4bfa191bd80/original=true/ab435bf4-414b-46e3-ac9c-d4bfa191bd80.jpeg" />Kofi: <a target="_blank" rel="ugc" href="https://ko-fi.com/mrweaz">https://ko-fi.com/mrweaz</a><hr /><span style="background-color:rgb(26, 27, 30);color:rgb(141, 141, 143);font-family:-apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, Helvetica, Arial, sans-serif, &quot;Apple Color Emoji&quot;, &quot;Segoe UI Emoji&quot;;font-size:16px">Disclaimer: I hold no liability for any images produced, including their legality, ethical implications, or potential misuse. By using this model, you agree that all responsibility for generated content rests solely with you. Use responsibly and within legal boundaries.

Minimax H3: EZ  | Turbo . Optimal . RTX Upscale . LTX Refine | R2VA . ITVA . T2VA . FL2VA

UPDATE COMFYUI TO LATEST BRANCH. NOT RELEASES. AND INSTALL NODES USING GIT CLONE.CHECK CONSOLE FOR MISSING PACKAGES LIKE SAGEATTN/TRITON. GIT PULL YOUR NODES, THEY UPDATE FAST.DO NOT USE THE MANAGER TO INSTALL THE NODES. YOU NEED THE LATEST UPDATES FROM GITHUB. THE UPDATES MOVE SUPER FAST SO THINGS WILL BREAK<ul><li>Reworked alot of it to be more streamlined.</li><li>Added FLF2V switch per request.</li><li>Added optimal boosters for fast iterations with minimal quality drops.</li><li>Added an optional LTX Refinement method (Can fix ghosting, sometimes hit and miss)</li><li>Fixed some issues.</li></ul>

Files with SHA256: fb731a20fc5821d9770f4b9545a8edcce21463e7849cd477f3a414db33d517e6