What GPU do I need to run genmo/mochi-1-preview?
A 10B-parameter video generation model. 10.0B parameters, published in F32. View on Hugging Face
mochi-1-preview is published by genmo on Hugging Face, with 7,745 downloads and 1,359 likes to date. It's a unlisted-architecture model built for text-to-video, published natively in F32.
What mochi-1-preview is
mochi-1-preview is a 10B-parameter video generation model published by Genmo on Hugging Face, released under Apache 2.0.
License note: fully permissive, no gating and no usage restrictions. Facts in this section are sourced from mochi-1-preview's Hugging Face model card, not benchmarked by Aquanode.
What it's used for
- Text-to-video / image-to-video generation
- Short-form creative video
VRAM required & cheapest live GPU fit
Required VRAM = weight size at each precision, plus a fixed overhead for activation memory and allocator fragmentation. Diffusion and video models carry no KV-cache. The real driver of extra memory is output resolution and frame count, which this flat overhead does not model. Full formula and assumptions: methodology.
| Precision | Weight size | Required VRAM | Cheapest live fit | GPUs needed | Est. $/hr (full fit) |
|---|---|---|---|---|---|
| FP32 | 37.4 GB | 44.8 GB | RTX A6000 | 1 | $0.363/hr |
| cheaper alt. | V100 | 3 | $0.264/hr | ||
| FP8 (quantized) | 9.3 GB | 11.2 GB | RTX 5060 Ti | 1 | $0.110/hr |
| INT4 (quantized) | 4.7 GB | 5.6 GB | RTX 5060 Ti | 1 | $0.110/hr |
A GPU is only matched to a row if its hardware supports that precision, and the primary recommendation is always a single-GPU fit when one exists.
INT4 requires a quantized checkpoint actually published for this model, check its Hugging Face page before relying on this row.
Cheapest way to run mochi-1-preview at its published (F32) precision: 1× RTX A6000, at $0.363/hr per GPU ($0.363/hr total). Quantizing to FP8 or INT4 (rows above) can cost less, but requires a compatible quantized checkpoint to exist for this model.
mochi-1-preview: common questions
Can mochi-1-preview run on a single GPU?
Yes, but not on a desktop card. At FP32 it needs 44.8 GB of VRAM, more than a single 32 GB desktop card holds. The cheapest live single-card fit is a 48.0 GB RTX A6000 at $0.363/hr.
Can mochi-1-preview run in 16-bit instead of FP32?
Yes. Its published weights are FP32, 37.4 GB, or 44.8 GB once inference overhead is added. Loading the same weights in 16 bits halves that to 18.7 GB, or 22.4 GB with overhead. That moves it onto a 24 GB card, which the FP32 weights do not fit. How much accuracy the cast costs is model-specific and is not measured here.
What is the least VRAM mochi-1-preview can run in?
5.6 GB, at INT4 (quantized), which fits a 6 GB card, against 44.8 GB at FP32. That row assumes an INT4 checkpoint has actually been published for this model, so check its Hugging Face page before planning around it.
Does quantizing mochi-1-preview lower the GPU bill?
Yes. At FP32 the cheapest live fit is one RTX A6000 at $0.363/hr. At FP8 (quantized) it drops to one RTX 5060 Ti at $0.110/hr, provided a quantized checkpoint exists for it.
How to run mochi-1-preview
Run mochi-1-preview with Diffusers (Python)
From genmo/mochi-1-preview's own model card.
import torch
from diffusers import MochiPipeline
from diffusers.utils import export_to_video
pipe = MochiPipeline.from_pretrained("genmo/mochi-1-preview")
# Enable memory savings
pipe.enable_model_cpu_offload()
pipe.enable_vae_tiling()
prompt = "Close-up of a chameleon's eye, with its scaly skin changing color. Ultra high resolution 4k."
with torch.autocast("cuda", torch.bfloat16, cache_enabled=False):
frames = pipe(prompt, num_frames=84).frames[0]
export_to_video(frames, "mochi.mp4", fps=30)Source: https://huggingface.co/genmo/mochi-1-preview/raw/main/README.md
Run mochi-1-preview with ComfyUI
Aquanode's ComfyUI template comes with ComfyUI preinstalled. Download genmo/mochi-1-preview's checkpoint into the models folder and load it in a workflow; this is a real Aquanode template, but loading this specific checkpoint is a manual step, not a one-click deploy.
Deploy mochi-1-preview on Aquanode
Aquanode has no one-click deploy template for mochi-1-preview; it comes with ComfyUI preinstalled, so you only need to load the checkpoint, not install anything. Aquanode sells GPU pods billed per second, not a hosted inference API.
- Launch the ComfyUI template sized to the requirement above (1× RTX A6000 or larger).
- Open a terminal on the pod, or save one of the commands above as a startup script so it runs automatically the first time the pod boots.
- Run the command and connect to the resulting endpoint.
Launch the ComfyUI templateWeight-to-VRAM math, the fit rules, and how live prices are normalized: full methodology.
Related reading: RTX A6000 pricing and specs, The NVIDIA Inception program, explained, Google Colab alternatives for dedicated GPU access, Free GPU credits for students and researchers, and RunPod volume disk vs. network volume, compared.
