Making a 40-Second Anime Episode Locally with ComfyUI on an RTX 3090
Melvin Vivas · X video post · 2026-10-09 · 0:40 · 1,022 views · Open on X
Topics: AI Dev Tools & Productivity, AI Agents, Tool Use & MCP, Industry Trends & Job Market · Level: intermediate
Summary
Melvin Vivas shows a short anime clip he made entirely on his own PC. Grok Bot acted as the "director" that coordinated the production, ComfyUI ran the generation workflows, and a single RTX 3090 did the work. Images came from Krea-2 Turbo (FP8) and video from MiniMax H3 (INT8) using Ref2VA. His point is that quantized open models now make it possible to produce multi-shot video on consumer hardware at no cost.
Key points
- The setup has three roles: Grok Bot is the director that coordinates production, ComfyUI is the production studio that runs the workflows, and an RTX 3090 (24 GB) is the hardware.
- Image generation used Krea-2 Turbo, quantized to FP8.
- Video generation used MiniMax H3, quantized to INT8, in Ref2VA mode. Ref2VA appears to be reference-to-video, which starts from a reference image.
- Quantization (FP8/INT8) is what lets large image and video models fit and run on a single 24 GB consumer GPU.
- Running locally means there are no per-generation API costs. The creator says it's 'free' because it runs on his PC.
- The finished result was a 40-second anime episode built from multiple generated clips.
- The workflow is to have an LLM assistant plan the shots and prompts, generate keyframe images, then animate them into video clips.
Resources mentioned
- ComfyUI · tool · x.com · free
The official X account for ComfyUI, the node-based tool for generative AI workflows, which announces things like ComfyUI Agent.
Also in: Grok as "Director": Making a Local Anime Clip with ComfyUI on an RTX 3090 (Melvin Vivas on X · notes), FLUX 3 Image Klein: Open-Weights Image Model Now Supported in ComfyUI (Melvin Vivas on X · notes), ComfyUI Templates: Chain Ready-Made Workflows Instead of Building From Scratch (Melvin Vivas on X · notes), Comfy Agent: An AI Agent That Builds ComfyUI Workflows in Comfy Cloud (Melvin Vivas on X · notes) and 32 more - Grok Bot · tool · x.ai · paid
An AI personal assistant the creator describes as having memory, day planning, task scheduling and the ability to do background work without any coding.
Also in: Turn repeated agent tasks into reusable Skills (Melvin Vivas on X · notes), Using a Grok Bot to find free AI credit giveaways (Melvin Vivas on X · notes), Grok Bot can now use phone calendar, contacts and reminders (Melvin Vivas on X · notes), Grok Bot Adds 1Password Integration for Agent Logins (Melvin Vivas on X · notes) and 129 more - Krea 2 Turbo · tool · huggingface.co · free
An image generation model, run here as an FP8-quantized version to make keyframe images locally.
Also in: Krea 2 Open Weights: Raw (Undistilled) vs Turbo (Distilled) Image Models (Melvin Vivas on X · notes) - MiniMax H3 · tool · huggingface.co · free
A MiniMax reference-to-video generation model, run here as an INT8 quantized version for local video head swapping.
Also in: Let an Agent Run Overnight Local Video Generation with ComfyUI (Melvin Vivas on X · notes), Free local anime video generation with Minimax H3 on an RTX 3090 (Melvin Vivas on X · notes), A day of work done from a phone with Grok Bot agents (Melvin Vivas on X · notes), Local $0 Video Head Swap: MiniMax H3 Ref2VA (INT8) + LoRA in ComfyUI (Melvin Vivas on X · notes) and 4 more - NVIDIA GeForce RTX 3090 · tool · nvidia.com · paid
A consumer GPU with 24 GB of VRAM, used as the local hardware for image and video generation.
Also in: Running an Overnight 100-Iteration Agent Loop with a Local Qwen Model (Melvin Vivas on X · notes), Grok as "Director": Making a Local Anime Clip with ComfyUI on an RTX 3090 (Melvin Vivas on X · notes), Local $0 Video Head Swap: MiniMax H3 Ref2VA (INT8) + LoRA in ComfyUI (Melvin Vivas on X · notes), Portable Computer Now Runs AI Agents and Models Locally on NVIDIA RTX PCs (Melvin Vivas on X · notes) and 6 more
Try this
- Make a short anime episode locally: have an LLM assistant act as director, writing the shot list and prompts; generate keyframes with Krea-2 Turbo (FP8) in ComfyUI; then animate them with MiniMax H3 (INT8) Ref2VA on a 24 GB GPU.
More in AI Dev Tools & Productivity
- llama.app: Run Local Models on a Mac Without the llama.cpp CLI
- Try Devin by Connecting Your ChatGPT Go, Plus or Pro Plan
- Coding from your phone: Grok Bot hands tasks to Cursor Cloud Agents
- Grok Bot 101 official guide
- Switch between Grok Bots with keyboard shortcuts
- Free local anime video generation with Minimax H3 on an RTX 3090