Mochi 1Unclaimed profile

Open-source text-to-video model that turns written prompts into short, high-motion video clips.

Task: Producing Video & AnimationUse case: Generate high quality videos from text prompts using state of the art video worlModels: mochi 1

0.0

based on 0 reviews

Are you the vendor?

reviews

Manzia generates an at-a-glance sentiment summary once an Agent has 5 approved reviews. This Agent has 0.

Summarized from 3 community tutorials · July 2026

Summarized from 3 community tutorials · July 2026

Local text-to-video generationRuns the open-weights Mochi 1 model entirely on local hardware via ComfyUI instead of a hosted cloud service.
ComfyUI node-based workflowWires up diffusion model loader, CLIP text encoder, and VAE (native or KJ wrapper nodes) to build a text-to-video pipeline.
Adjustable generation settingsPrompt, negative prompt, resolution, frame length, sampler steps, seed, and precision (fp8/fp16) are all configurable per run.
Quantized checkpoints for consumer GPUsfp8-scaled all-in-one checkpoint trims VRAM needs so the model can run on cards like an RTX 4090 instead of requiring an H100.
VAE tiling for longer clipsEnabling tiled VAE decode is needed to generate multi-second clips without the decode step running out of memory.

Summarized from 3 community tutorials · July 2026