~/runthismodel
daemon okbuild 5a3c91d00:00:00Z
./models/browse/mochi-1-preview
Genmo · video-gen
Mochi 1 Preview
Apache-licensed text-to-video. Strong motion, 5.4-second clips. ~30 GB VRAM at FP16.
10b paramsasymditapache-2.03030 GB vram
about·model card

Mochi 1 from Genmo offers some of the most cinematic motion in the open-source video space. Apache-2.0 licensing makes it commercial-friendly — a rarity for video models. Plan for 30+ GB VRAM at FP16.

probe://hardware·which quants fit your rig
we auto-detect via WebGL/WebGPU. select manually if your GPU isn't recognized.
./quantizations·1 variants
QuantizationBitsFile SizeVRAM NeededRAM NeededQuality
FP161620 GB30 GB32 GB
100%

How to run Mochi 1 Preview

Pick a runtime — copy & paste. Commands are pre-filled with this model’s repo.

GUI. Browse → download → chat. MLX on Apple Silicon.

LM Studio home →
  1. 1

    Open LM Studio

    Go to the 🔍 Search tab.

  2. 2

    Search for

    genmo/mochi-1-preview
  3. 3

    Download

    Pick the FP16 quant — best balance of size vs. quality.

  4. 4

    Chat

    Hit ▶ Load Model and start chatting. Toggle 'Local Server' to expose an OpenAI-compatible API on :1234.

Community benchmarks

Real seconds-per-image reports from people running Mochi 1 Preview on actual hardware.

No community runs yet for this model. Be the first to submit your numbers.

faq·common questions
how much VRAM do I need to run Mochi 1 Preview?

Mochi 1 Preview requires 30 GB VRAM minimum with FP16 quantization. For full precision you need 30 GB.

which quant should I pick?

Q4_K_M is the best quality/VRAM balance — ~92% of FP16 quality at ~25% the footprint. Q8_0 is near-lossless if you have the headroom.

faq://ai-curated·20 entries
What GPU do I need to run Mochi 1 Preview?

To run Mochi 1 Preview, you need a GPU with at least 30 GB of VRAM, such as an NVIDIA A6000 or RTX 3090.

Is Mochi 1 Preview good for coding?

Mochi 1 Preview is primarily designed for generating short video clips and may not be optimized for coding tasks, which typically require different model architectures.

Mochi 1 Preview vs Llama 3.1 8B?

Mochi 1 Preview has 10 billion parameters and focuses on video generation, while Llama 3.1 8B is a smaller language model with 8 billion parameters, better suited for text-based tasks.

Can I run Mochi 1 Preview on a Mac?

Running Mochi 1 Preview on a Mac is possible if your Mac has a compatible GPU with at least 30 GB of VRAM, which is rare in current Mac models.

How much VRAM does Mochi 1 Preview need?

Mochi 1 Preview requires 30 GB of VRAM to run at FP16 precision, which is necessary for optimal performance and quality.

Is Mochi 1 Preview censored?

Mochi 1 Preview is not inherently censored, but its outputs are governed by the training data and the model's architecture, which may include content filters.

Is Mochi 1 Preview commercial-use allowed?

Yes, Mochi 1 Preview is licensed under the Apache-2.0 license, allowing for both personal and commercial use without restrictions.

Mochi 1 Preview context length?

The context length for Mochi 1 Preview is currently unknown, but it is designed to generate 5.4-second video clips from text inputs.

Does Mochi 1 Preview support function calling?

Mochi 1 Preview is primarily designed for video generation and does not natively support function calling like some language models.

Mochi 1 Preview quantization options?

Mochi 1 Preview supports quantization, but the exact options and impact on VRAM usage and performance are not specified; typically, FP16 is recommended for best results.

Can Mochi 1 Preview run on CPU?

While Mochi 1 Preview can technically run on a CPU, it is highly inefficient and not recommended due to the high computational demands and the requirement of 30 GB of VRAM.

Mochi 1 Preview fine-tuning?

Mochi 1 Preview can be fine-tuned, but this requires significant computational resources and expertise in training large models.

Mochi 1 Preview system requirements?

Mochi 1 Preview requires a powerful GPU with at least 30 GB of VRAM, ample RAM, and a modern CPU to handle the computational load efficiently.

Mochi 1 Preview performance benchmark?

Performance benchmarks for Mochi 1 Preview are not widely available, but it is known to generate 5.4-second video clips with strong motion, requiring significant VRAM and processing power.

Mochi 1 Preview for RAG?

Mochi 1 Preview is not optimized for Retrieval-Augmented Generation (RAG) tasks, which typically involve text retrieval and synthesis rather than video generation.

Mochi 1 Preview for agents?

Mochi 1 Preview is not designed for creating conversational agents or chatbots; it is specialized for generating short video clips from text.

Mochi 1 Preview for coding vs general?

Mochi 1 Preview is not suitable for coding tasks or general-purpose text generation; it is specifically designed for generating short video clips.

Mochi 1 Preview vs ChatGPT?

Mochi 1 Preview is a 10B parameter model focused on video generation, while ChatGPT is a larger language model designed for text-based interactions and conversational tasks.

Mochi 1 Preview download size?

The download size for Mochi 1 Preview is not explicitly stated, but given its 10 billion parameters, the model file is likely to be several gigabytes in size.

Best quant for Mochi 1 Preview?

For Mochi 1 Preview, using FP16 quantization is generally recommended as it balances performance and quality while requiring 30 GB of VRAM.