minimax-studio
Coordinates GPU VRAM with a local Ollama server: queries the Ollama API to detect loaded LLMs and signals them to unload (keep_alive: 0) before video generation, and exposes a tool (h3_gpu_vacate) to force Ollama to vacate VRAM on demand so ComfyUI has full memory for MiniMax H3 inference.
This server does not currently have a deployable release on Glama. Follow the repository's setup instructions to install and configure it for your MCP client, including any required credentials and local dependencies.
MiniMax H3 Studio (Local Omni-Modal Video & Audio)
Local control panel and MCP server for MiniMax H3, running 100% locally on this machine via ComfyUI and AMD ROCm (Radeon RX 9060 XT 16GB VRAM).
Unlike Google Flow or cloud video generators, MiniMax H3 Studio:
Generates synchronized video AND audio simultaneously in a single omni-modal pass.
Costs 0 credits and has no usage limits or cloud queues.
Features an Automated GPU Synchronizer & VRAM Arbiter that automatically manages VRAM between Ollama, ComfyUI, Open WebUI, and other apps with zero memory conflicts (
hipErrorOutOfMemoryprevention).Exposes native MCP tools to Claude Code, Antigravity (
agy), Codex, and Gemini CLI.
Assistant (Claude, agy, Codex) ──stdio──> minimax-studio ──REST / WebSocket──> ComfyUI (Local AMD ROCm)
│ │
└── GPU Synchronizer & Arbiter ────────┴──> Ollama VRAM Coordination🚀 Quick Start
1. Launch the Local Web Studio
cd /home/bakhtiarrifai/projects/minimax-studio
npm run studio # Opens Web UI at http://127.0.0.1:87952. Drive from AI Assistants (Claude Code, Antigravity, Codex)
The MCP server is already registered globally:
In Antigravity: Type
h3_statusor ask "Generate a 2-second clip of a cyberpunk city with orchestral audio".In Claude Code: "Use the minimax tool to queue a video generation".
In Codex: Use
h3_generateandh3_wait.
To re-register at any time:
npm run registerRelated MCP server: comfy-ae-video-factory-mcp
🎬 MiniMax H3 Generation Capabilities
Text-to-Video (T2V) + Audio:
Visuals generated by the 33.1B DiT model (
minimax_h3_fl2va_pruned_fp8_scaled.safetensors).Text conditioning parsed by Qwen3-VL 32B (
qwen3vl_32b_minimax_h3_nvfp4_awq.safetensors).Native soundscape decoded by Audio VAE (
minimax_h3_audio_vae_fp32.safetensors).
First-to-Last Frame (I2V / FL2V):
Provide
first_frame_pathorlast_frame_pathto smoothly animate between still images.
Cinematic Techniques (
h3_techniques):Over 40 built-in camera movements (360 Orbit, Dolly In/Out, FPV Drone, Bullet Time, Tracking Shot).
Visual styles (Golden Hour, Cyberpunk Neon, 35mm Film, Noir, Volumetric Fog).
Audio styles (Cinematic Orchestral, Nature Foley, Synthwave, Sci-Fi Atmosphere).
FFmpeg Movie Stitcher (
h3_cut):Seamlessly stitches multiple generated shots into a single finished MP4 film.
Mixes or replaces audio tracks with custom music or voiceovers.
⚡ Server GPU Synchronization & Conflict-Free Sequencing
The server hosts multiple AI services sharing an AMD Radeon RX 9060 XT (16 GB VRAM):
Ollama (
gemma4-12b-64k, Open WebUI, Telegram Bot)ComfyUI (MiniMax H3, Wan2.1)
How comfy_gpu_sync Coordinates Seamlessly:
Pre-Run Coordination: Whenever MiniMax H3 Studio or ComfyUI receives a generation job, the synchronizer queries Ollama (
GET http://127.0.0.1:11434/api/ps). If an LLM is in VRAM, it signals Ollama to unload (keep_alive: 0), releasing ~10.2 GB VRAM in under 0.5s.Pre-VAE Memory Guard (
ComfyUnloadModelForVAE): Automatically unloads the 33.1B DiT model from VRAM immediately after latent sampling finishes, guaranteeing that the Video and Audio VAE decoders have full 16 GB memory address space withouthipErrorInvalidValue.Idle Auto-Release: Once ComfyUI generation queue is empty for 60 seconds, ComfyUI unloads models and clears PyTorch ROCm cache.
On-Demand LLM Reloading: When a user or Telegram bot asks Ollama a question, Ollama reloads the LLM into full VRAM instantly with zero errors.
🛠️ MCP Tools Reference
Tool | Description |
| Check GPU VRAM, queue status, active jobs, and model readiness. |
| Queue a video + audio generation with prompt, techniques, duration, and resolution. |
| Wait for job completion with live progress percentage and get the output video path. |
| Join multiple clips into a final film with local FFmpeg and optional music mix. |
| Catalog of 40+ cinematic camera movements, lighting looks, and sound presets. |
| Manually force Ollama or ComfyUI to vacate VRAM on demand. |
🐧 Linux Server & AMD Radeon RX 9060 XT (16GB) Support
MiniMax H3 Studio was built and battle-tested on this Linux server with an AMD Radeon RX 9060 XT (16 GB VRAM) running ROCm 7.2.
For the complete technical setup, memory guard architecture, and ROCm configuration, read: 👉 LINUX_ROCM_RX9060XT_GUIDE.md
Server Highlights:
DiT FP8 Scaled: Runs 33.1B diffusion transformer directly on RX 9060 XT RDNA 4 hardware.
CPU Offloading: Qwen3-VL 32B text encoder offloaded to CPU to keep 100% of 16 GB VRAM free for video generation.
Pre-VAE Memory Guard: Automatically unloads DiT before VAE decode to prevent ROCm
hipErrorInvalidValue.Automated Multi-Service Sync: Coordinates with Ollama and other apps via
comfy_gpu_sync.
🪟 Windows & NVIDIA RTX 3060 (12GB) Support
MiniMax H3 Studio is fully compatible with Windows 10/11 and NVIDIA GeForce RTX 3060 (12 GB VRAM).
For complete step-by-step setup in Indonesian & English, read: 👉 WINDOWS_RTX3060_GUIDE.md
One-Click Windows Scripts:
setup_windows.bat: Installs npm dependencies and copiescomfy_gpu_syncinto ComfyUI.download_models_windows.bat: Downloads the 4 quantized models from Hugging Face Hub.start_comfyui_rtx3060.bat: Starts ComfyUI with--lowvramflag for RTX 3060.start_studio.bat: Launches the Web Studio control panel athttp://127.0.0.1:8795.
📂 Project Structure
minimax-studio/
├── setup_windows.bat # One-click Windows setup script
├── download_models_windows.bat# Automated Hugging Face model downloader for Windows
├── start_comfyui_rtx3060.bat # ComfyUI launcher optimized for RTX 3060 12GB
├── start_studio.bat # Web Studio launcher for Windows
├── LINUX_ROCM_RX9060XT_GUIDE.md # Complete Linux ROCm + RX 9060 XT guide
├── WINDOWS_RTX3060_GUIDE.md # Complete Windows + RTX 3060 guide
├── LICENSE # MIT License
├── package.json # MCP server & Studio scripts
├── custom_nodes/
│ └── comfy_gpu_sync/ # ComfyUI VRAM arbiter & auto-sync custom node
├── src/
│ ├── config.js # Ports, paths, and default settings
│ ├── gpu_sync.js # Cross-platform GPU VRAM arbiter (AMD ROCm & NVIDIA CUDA)
│ ├── comfy_client.js # ComfyUI WebSocket & REST MiniMax H3 pipeline
│ ├── techniques.js # 40+ cinematic camera & audio presets
│ ├── cut.js # Cross-platform FFmpeg video concatenation
│ ├── server.js # HTTP & WebSocket API server + scenes gallery
│ ├── studio.js # Web Studio launcher CLI
│ ├── mcp.js # MCP Server (stdio)
│ └── register.js # AI assistant registration script
├── studio/
│ └── index.html # Sleek dark-mode Studio Web UI with live player
└── scripts/
├── download_models_linux.sh # Hugging Face model downloader for Linux
├── download_models.py # Cross-platform Hugging Face model downloader (Python)
├── generate_full_episode_13.js # Automated 4-scene episode production
└── test-connection.js # Health check & integration test📄 License
MIT License. See LICENSE for details.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
This server cannot be deployed
Maintenance
Related MCP Connectors
Transform video, audio and images, and generate media from prompts. FFmpeg, captions, models.
FFmpeg as a service for AI agents: typed video editing tools, async jobs, downloadable outputs.
Generate, extend and edit AI images and videos, then assemble clips on a timeline and render.
LLM chat, text tools, image generation, editing, batch image jobs, and asynchronous video generation
Related MCP Servers
- AlicenseAqualityDmaintenanceEnables video generation through OpenAI's Sora 2 API, allowing users to create, monitor, and manage AI-generated videos. It also provides tools for merging video clips and creating fade animations from static images using FFmpeg.67 npm2MIT
- AlicenseBqualityDmaintenanceEnables AI video automation pipeline: ComfyUI image-to-video, FFmpeg processing, After Effects template rendering, review, and multi-platform publishing preparation.9MIT
- FlicenseBqualityDmaintenanceEnables AI assistants to perform video and audio processing tasks such as format conversion, cutting, merging, and adding effects via FFmpeg, with hardware acceleration support.193-
- AlicenseAqualityBmaintenanceEnables agents to generate images, video, and 3D meshes fully offline on the local GPU through a shared on-demand ComfyUI backend, with additional tools for retopology, texturing, rigging, background removal, upscaling, and speech.16MIT