Video catalogue · Verified local path

Vimoo 1.0 Lite local guide

Apache-licensed local text-to-video and image-to-video family with Diffusers, ComfyUI and Apple Silicon MLX paths.

Choose an app

Start with the recommended app, or choose another supported option.

Compare video models
Open model filesRecommended verified starting pointOpen in UnslothLaunches Unsloth Desktop on this modelGet ComfyUI DesktopOfficial desktop app for this compatible modelOpen Diffusers setupAdvanced Python runtimeCheck my machineSee whether this model fits your saved hardware

Desktop app links require the app to be installed. If nothing opens, LocalClaw will show app-download and model-file fallbacks.

What it does

Text To VideoImage To VideoAnimationFirst Frame Video

Vimoo AI publishes Apache-2.0 source repositories for the Diffusers examples, ComfyUI nodes and MLX runtime, plus public Hugging Face safetensors for Vimoo 1.0 and the practical 9.5B Lite TI2V variants. The MLX path runs Vimoo-1.0-Lite TI2V Distilled on macOS 14+ Apple Silicon with Python 3.11, `hf download Vimoo-AI/Vimoo-1.0-Lite-MLX` and `vimoo-mlx` commands for 1280 x 720, 121-frame MP4 T2V or I2V output. The ComfyUI integration requires ComfyUI 0.34+, Python 3.10+, PyTorch 2.9.0 CUDA 13.0 and NVIDIA CUDA, ships template workflows and recommends dynamic VRAM, async offload, batch 1 and tiled VAE decoding for 16 GB VRAM. LocalClaw records 32 GB RAM and 16 GB accelerator memory as the conservative local floor for Lite or distilled workflows, with larger memory recommended for full 11.7B Vimoo 1.0 checkpoints.

Hardware figures are practical entry floors, not performance guarantees. Resolution, duration, precision, offloading and runtime versions can materially change memory use.

Strengths

  • Official Apache-2.0 weights and first-party runtime repositories
  • Text-to-video and image-to-video in Diffusers, ComfyUI and Apple Silicon MLX
  • Lite distilled path exports 5-second 720p MP4 clips without calling the hosted API

Limits to know

  • Hosted prompt enhancement and 1080p enhancer are API-side, not local inference
  • Best ComfyUI path currently targets CUDA 13.0 and recent NVIDIA drivers
  • English prompts are the validated path in the official release notes