Genmo

aid: genmo-ai
url: https://raw.githubusercontent.com/api-evangelist/genmo-ai/refs/heads/main/apis.yml
name: Genmo
kind: company
description: 'Genmo is a San Francisco research lab building open-source video generation models. Its flagship release, Mochi
  1, is an Apache-2.0 licensed 10B-parameter text-to-video diffusion model (AsymmDiT architecture) with strong prompt adherence
  and high-fidelity motion. Genmo does NOT publish a first-party hosted developer REST API. Programmatic access to Mochi comes
  in three honest forms: (1) the open-source model itself, self-hosted via the repo''s Python pipeline ("simple composable
  API"), CLI, Gradio UI, or ComfyUI; (2) third-party hosted inference APIs from partners such as Replicate and fal.ai; and
  (3) a credit-based consumer web app and Playground at genmo.ai/play. The endpoints documented here for hosted access are
  modeled from partner documentation and are not operated by Genmo.'
deliveryModel:
  model: saas
  open_source: false
  commercial: true
  callable_host: false
  label: Hosted service · you call their endpoint
  confidence: medium
  source:
  - pricing
  generated: '2026-08-28'
  method: derived
accessModel:
  pricing: freemium
  onboarding: unknown
  trial: false
  try_now: false
  public: false
  label: Freemium
  confidence: medium
  source:
  - plans
  generated: '2026-07-22'
  method: derived
image: https://kinlane-images.s3.amazonaws.com/shared/apis-json/icons/genmo-ai.png
tags:
- Video Generation
- AI Video
- Generative AI
- Text-to-Video
- Open Source
- Mochi
- Diffusion Model
tags_raw:
- Video Generation
- AI Video
- Generative AI
- Text-to-Video
- Open Source
- Mochi
- Diffusion Model
- Open-Source
created: '2026-07-11'
modified: '2026-07-25'
specificationVersion: '0.23'
apis:
- aid: genmo-ai:genmo-ai-mochi-replicate-api
  name: Mochi 1 on Replicate (Third-Party Hosted)
  tags:
  - Video Generation
  - Text-to-Video
  - Hosted Inference
  - Third Party
  tags_raw:
  - Video Generation
  - Text-to-Video
  - Hosted Inference
  - Third Party
  image: https://kinlane-images.s3.amazonaws.com/shared/apis-json/apis-json-logo.jpg
  humanURL: https://replicate.com/genmoai/mochi-1
  baseURL: https://api.replicate.com/v1
  properties:
  - url: https://replicate.com/genmoai/mochi-1
    type: Documentation
  - url: https://replicate.com/genmoai/mochi-1/api
    type: APIReference
  endpointsModeled: true
  description: Third-party hosted access to Genmo's Mochi 1 via Replicate's predictions API (model identifier genmoai/mochi-1).
    Runs on Nvidia H100 hardware, ~$0.42 per run, typically completing within 5 minutes at 480p. Invoked through Replicate's
    standard async prediction pattern - POST a prediction with the model version and inputs, poll or receive a webhook, then
    fetch the output video URL. This API is operated by Replicate, not Genmo; endpoints are modeled from Replicate's platform
    documentation.
- aid: genmo-ai:genmo-ai-mochi-fal-api
  name: Mochi v1 on fal.ai (Third-Party Hosted)
  tags:
  - Video Generation
  - Text-to-Video
  - Hosted Inference
  - Third Party
  tags_raw:
  - Video Generation
  - Text-to-Video
  - Hosted Inference
  - Third Party
  image: https://kinlane-images.s3.amazonaws.com/shared/apis-json/apis-json-logo.jpg
  humanURL: https://fal.ai/models/fal-ai/mochi-v1
  properties:
  - url: https://fal.ai/models/fal-ai/mochi-v1/api
    type: APIReference
  endpointsModeled: true
  description: Third-party hosted access to Mochi via fal.ai (model identifier fal-ai/mochi-v1), using fal's queue API - submit
    with prompt, negative_prompt, seed, enable_prompt_expansion, and num_frames, then check status and fetch the resulting
    video URL. NOTE - as of the review date fal.ai lists this endpoint as deprecated / no longer supported. This API is operated
    by fal.ai, not Genmo; endpoints are modeled from fal's documentation and retained for discoverability.
- aid: genmo-ai:genmo-ai-mochi-pipeline-api
  name: Genmo Mochi Pipeline (Self-Hosted OSS)
  tags:
  - Video Generation
  - Text-to-Video
  - Open Source
  - Self-Hosted
  tags_raw:
  - Video Generation
  - Text-to-Video
  - Open Source
  - Self-Hosted
  - Open-Source
  image: https://kinlane-images.s3.amazonaws.com/shared/apis-json/apis-json-logo.jpg
  humanURL: https://github.com/genmoai/mochi
  properties:
  - url: https://github.com/genmoai/mochi
    type: Documentation
  - url: https://huggingface.co/genmo/mochi-1-preview
    type: SourceCode
  - url: https://www.genmo.ai/blog/mochi-1-a-new-sota-in-open-text-to-video
    type: Blog
  endpointsModeled: true
  description: The first-party access surface is the open-source model itself. The Mochi repo (Apache-2.0) ships a "simple
    composable API" in Python - MochiSingleGPUPipeline built from T5ModelFactory, DitModelFactory, and DecoderModelFactory
    - invoked with prompt, resolution, num_frames, inference steps, guidance scale, and seed to generate 480p video locally.
    It also provides a cli.py, a Gradio UI (gradio_ui.py), and native ComfyUI support. This is a code library / local inference
    pipeline, not a hosted network API; weights (~60GB VRAM) are downloaded from Hugging Face. There is no Genmo REST endpoint
    - self-host it or use a partner (see the Replicate and fal.ai entries).
common:
- type: GitHubOrganization
  url: https://github.com/genmoai
- type: LinkedIn
  url: https://www.linkedin.com/company/genmoai
- type: Website
  url: https://www.genmo.ai
- type: Documentation
  url: https://github.com/genmoai/mochi
- type: Playground
  url: https://www.genmo.ai/play
- type: Plans
  url: plans/genmo-ai-plans-pricing.yml
- url: https://www.genmo.ai/blog
  type: Blog
maintainers:
- FN: Kin Lane
  email: kin@apievangelist.com