{
 "slug": "cogvideo",
 "name": "CogVideo / CogVideoX",
 "author": "THUDM",
 "density": "full",
 "status": "active",
 "status_note": null,
 "alternatives_page": null,
 "one_liner": "Open-source text-to-video and image-to-video generation models (CogVideo, CogVideoX) with inference and fine-tuning code.",
 "category": {
  "slug": "ai-video",
  "name": "AI video generators and platforms"
 },
 "tags": [
  "Open source"
 ],
 "styles": [],
 "logo_url": "https://agentgid.com/media/avatars/zai-org.png",
 "logo_letter": "C",
 "links": {
  "repo": "https://github.com/zai-org/CogVideo",
  "website": null,
  "api_docs": null
 },
 "preview": null,
 "verdicts": [
  {
   "key": "price",
   "label": "Price",
   "value": "Free",
   "status": "yes",
   "sub": "No paid plan",
   "quote": null,
   "note": "Free open-source code; models can also be tried online in Hugging Face and ModelScope Spaces.",
   "source": {
    "label": "github.com/zai-org/CogVideo",
    "url": "https://github.com/zai-org/CogVideo"
   },
   "n": 1
  },
  {
   "key": "commercial",
   "label": "Commercial use",
   "value": "With conditions",
   "status": "cond",
   "sub": "Apache-2.0",
   "quote": "The CogVideoX-5B model (Transformers module, include I2V and T2V) is released under",
   "note": "2B model and code are Apache 2.0; 5B models need the free commercial registration and stay under 1 million visits a month.",
   "source": {
    "label": "github.com/zai-org/CogVideo",
    "url": "https://github.com/zai-org/CogVideo#model-license"
   },
   "n": 2
  },
  {
   "key": "credit",
   "label": "Credit the author",
   "value": "Not needed",
   "status": "yes",
   "sub": "Keep the notice in code copies",
   "quote": "You must give any other recipients of the Work or Derivative Works a copy of this License",
   "note": "No credit on screen. Keep the license notice inside the code you ship.",
   "source": {
    "label": "LICENSE on GitHub",
    "url": "https://github.com/THUDM/CogVideo/blob/HEAD/LICENSE"
   },
   "n": 3
  },
  {
   "key": "agent_access",
   "label": "Agent access",
   "value": "Download",
   "status": "info",
   "sub": "Files by hand",
   "quote": null,
   "note": "There is no package or API. Files are downloaded by hand from the official site.",
   "source": {
    "label": "github.com/zai-org/CogVideo",
    "url": "https://github.com/zai-org/CogVideo"
   },
   "n": 4
  }
 ],
 "license": {
  "name": "Apache-2.0 (code and CogVideoX-2B); CogVideoX LICENSE (CogVideoX-5B)",
  "url": "https://github.com/zai-org/CogVideo/blob/HEAD/LICENSE",
  "summary": "The code and the small 2B model are Apache 2.0, free for paid products. The 5B models use their own license, which is free for research. Commercial use needs a free sign-up and is capped at one million visits a month.",
  "pages_read": 3,
  "rules": [
   "CogVideoX-5B weights follow the separate CogVideoX LICENSE on Hugging Face; read it before commercial use. The code and CogVideoX-2B are Apache-2.0.",
   "Keep a copy of the Apache 2.0 license and the copyright and NOTICE files when you redistribute the code.",
   "Mark modified files with a prominent notice that you changed them.",
   "The README says only the CogVideoX-2B weights are Apache 2.0, so check the model license for the other weights (e.g. 5B) before commercial use.",
   "Do not use THUDM trade names or trademarks beyond describing where the work came from."
  ],
  "rows": [
   {
    "label": "Commercial use",
    "value": "With conditions",
    "status": "cond",
    "quote": "The CogVideoX-5B model (Transformers module, include I2V and T2V) is released under",
    "source": {
     "label": "github.com/zai-org/CogVideo",
     "url": "https://github.com/zai-org/CogVideo#model-license"
    },
    "note": null
   },
   {
    "label": "Credit",
    "value": "Not needed",
    "status": "yes",
    "quote": "You must give any other recipients of the Work or Derivative Works a copy of this License",
    "source": {
     "label": "LICENSE on GitHub",
     "url": "https://github.com/THUDM/CogVideo/blob/HEAD/LICENSE"
    },
    "note": null
   },
   {
    "label": "Inside your app or game",
    "value": "With conditions",
    "status": "cond",
    "quote": "The CogVideoX-5B model (Transformers module, include I2V and T2V) is released under",
    "source": {
     "label": "github.com/zai-org/CogVideo",
     "url": "https://github.com/zai-org/CogVideo#model-license"
    },
    "note": null
   },
   {
    "label": "Pass the files on to your users",
    "value": "With conditions",
    "status": "cond",
    "quote": "Under the terms and conditions of this license, the licensor hereby grants you a non-exclusive, worldwide, non-transferable, non-sublicensable, revocable, royalty-free copyright license.",
    "source": {
     "label": "huggingface.co",
     "url": "https://huggingface.co/THUDM/CogVideoX-5b/raw/main/LICENSE"
    },
    "note": "Apache 2.0 parts may be passed on with a copy of the license; the 5B license is non-transferable and non-sublicensable."
   },
   {
    "label": "AI training",
    "value": "Not stated",
    "status": "unknown",
    "quote": null,
    "source": null,
    "note": null
   },
   {
    "label": "Bots and bulk downloads",
    "value": "Not stated",
    "status": "unknown",
    "quote": null,
    "source": null,
    "note": null
   }
  ],
  "checked_at": "2026-11-10",
  "commercial_use_page": "https://agentgid.com/assets/cogvideo/commercial-use/"
 },
 "overview": {
  "heading": "CogVideoX: short AI video clips on your own machine",
  "body": "CogVideoX is a family of free AI models that make short video clips from a text description or from a still picture. You run it yourself on a computer or rented server with a strong graphics card, so no outside video service is needed. The small model is free for any use; the larger ones need a free sign-up before commercial use."
 },
 "uses": [
  {
   "title": "A clip from a text description",
   "text": "Describe a scene in English, such as a panda playing guitar in a forest. You get a clip of about six seconds to use as a draft or a short loop.",
   "idea": false,
   "image": null
  },
  {
   "title": "Bring a still picture to life",
   "text": "Start from your own photo or drawing and add a short description of the motion; the picture becomes the first frame of the clip.",
   "idea": false,
   "image": null
  },
  {
   "title": "Turn a short idea into a full prompt",
   "text": "The models work best with long, detailed descriptions. A helper script lets a chat AI expand a one-line idea into such a description first.",
   "idea": false,
   "image": null
  },
  {
   "title": "Teach it your own style",
   "text": "Train the model a little further on your own short clips, for example product shots or interior scenes, so new clips follow that look.",
   "idea": false,
   "image": null
  },
  {
   "title": "Moving backgrounds for a landing page",
   "text": "Make a few calm, wordless clips, such as clouds or water, and play them muted behind a page header instead of buying stock footage.",
   "idea": true,
   "image": null
  },
  {
   "title": "Quick previews for a video script",
   "text": "Turn each scene of an ad or explainer script into a rough clip, so the team can judge the order and mood before real filming.",
   "idea": true,
   "image": null
  }
 ],
 "start": {
  "prompt": "Write a Python script that uses diffusers CogVideoXPipeline with THUDM/CogVideoX-2b (Apache 2.0) to turn a long English prompt into a 49-frame MP4 at 8 fps, with CPU offload and VAE tiling turned on to save graphics memory.",
  "install": "pip install --upgrade transformers accelerate diffusers imageio-ffmpeg",
  "install_note": "Use a machine with an NVIDIA graphics card and Python 3.10 to 3.12, then install the libraries the model card lists.",
  "code": [
   {
    "tab": "Step 1",
    "intro": "Load the model by its name and give it a long English description; the first run downloads the model files.",
    "source": "pipe = CogVideoXPipeline.from_pretrained(\n    \"THUDM/CogVideoX-5b\",\n    torch_dtype=torch.bfloat16\n)"
   }
  ]
 },
 "mistakes": [
  {
   "dont": "Asking for any number of frames, such as 60",
   "fix": "Use 8N+1 frames (default 49) for CogVideoX-2B/5B and 16N+1 (default 81) for CogVideoX1.5",
   "why": "The model table fixes the frame counts per model family.",
   "dont_code": false,
   "fix_code": false
  },
  {
   "dont": "Passing a short or non-English prompt straight to the model",
   "fix": "Expand the idea into a long English description first (convert_demo.py or any chat model)",
   "why": "The models were trained on long English prompts.",
   "dont_code": false,
   "fix_code": false
  },
  {
   "dont": "Keeping enable_sequential_cpu_offload() on when running on several GPUs",
   "fix": "Turn that option off for multi-GPU runs; keep it for single small GPUs",
   "why": "The README says that optimization must be disabled for multi-GPU inference.",
   "dont_code": false,
   "fix_code": false
  },
  {
   "dont": "Running the 2B model in BF16 like the 5B models",
   "fix": "Use torch.float16 for CogVideoX-2B and torch.bfloat16 for the 5B models",
   "why": "Each model should run in the precision it was trained in.",
   "dont_code": false,
   "fix_code": false
  },
  {
   "dont": "Shipping a commercial product on CogVideoX-5B because \"the code is Apache 2.0\"",
   "fix": "Use CogVideoX-2B for Apache 2.0 terms, or register for the 5B commercial license first",
   "why": "Only the code and the 2B weights are Apache 2.0; the 5B weights have their own license.",
   "dont_code": false,
   "fix_code": false
  }
 ],
 "agent": {
  "self_serve": false,
  "self_serve_text": null,
  "install": null,
  "repo": "https://github.com/zai-org/CogVideo",
  "api_docs": null,
  "api_account": "Not needed. Everything runs on your side.",
  "get_files": null,
  "formats": [],
  "limits": null,
  "mcp": null,
  "mcp_url": null,
  "credit_line": "Video made with CogVideoX (THUDM / Zhipu AI)",
  "rules_text": "- CogVideo / CogVideoX (AI video generators and platforms). Open-source text-to-video and image-to-video generation models (CogVideo, CogVideoX) with inference and fine-tuning code. Price: free; commercial use with conditions; no credit needed; ship inside your app with conditions.\n  Connect: download from https://github.com/zai-org/CogVideo (no API).\n  Rule: CogVideoX-5B weights follow the separate CogVideoX LICENSE on Hugging Face; read it before commercial use. The code and CogVideoX-2B are Apache-2.0.\n  Rule: Keep a copy of the Apache 2.0 license and the copyright and NOTICE files when you redistribute the code.\n  Rule: Mark modified files with a prominent notice that you changed them.\n  Rule: The README says only the CogVideoX-2B weights are Apache 2.0, so check the model license for the other weights (e.g. 5B) before commercial use.\n  Rule: Do not use THUDM trade names or trademarks beyond describing where the work came from.\n  License: https://github.com/zai-org/CogVideo/blob/HEAD/LICENSE (checked 2026-11-10; details: https://agentgid.com/assets/cogvideo/)",
  "agent_md": "https://agentgid.com/assets/cogvideo/agent.md",
  "card_json": "https://agentgid.com/assets/cogvideo/card.json"
 },
 "popularity": {
  "gid_score": 36,
  "stars": 13214,
  "downloads_7d": null,
  "downloads_7d_label": null,
  "downloads_change_4w_pct": null,
  "eco": null,
  "package": null,
  "repo": "zai-org/CogVideo",
  "agent_projects": null,
  "agent_projects_total": null,
  "downloads_unconfirmed": false,
  "series": {
   "downloads_weekly": [],
   "stars": [
    [
     "2026-11-07",
     13186
    ],
    [
     "2026-11-08",
     13186
    ],
    [
     "2026-11-09",
     13189
    ],
    [
     "2026-11-10",
     13190
    ],
    [
     "2026-11-11",
     13194
    ],
    [
     "2026-11-12",
     13197
    ],
    [
     "2026-11-13",
     13199
    ],
    [
     "2026-11-14",
     13201
    ],
    [
     "2026-11-15",
     13206
    ],
    [
     "2026-11-16",
     13208
    ],
    [
     "2026-11-17",
     13208
    ],
    [
     "2026-11-18",
     13211
    ],
    [
     "2026-11-19",
     13214
    ]
   ]
  },
  "stars_flat_since": null,
  "updated_at": "2026-11-19"
 },
 "alternatives": [
  {
   "slug": "cogvideo",
   "name": "CogVideo / CogVideoX",
   "why": null,
   "current": true,
   "price": "Free",
   "commercial": {
    "value": "With conditions",
    "status": "cond"
   },
   "access": "Download"
  },
  {
   "slug": "wan",
   "name": "Wan 2.2",
   "why": "Pick Wan 2.2 for a newer open model family that also covers speech-to-video and character animation at 480p and 720p.",
   "current": false,
   "price": "Free",
   "commercial": {
    "value": "Commercial OK",
    "status": "yes"
   },
   "access": "Download"
  },
  {
   "slug": "ltx",
   "name": "LTX-2 (LTX-2.5)",
   "why": "Pick LTX-2 when you want clips with sound and a ready ComfyUI setup on your own machine.",
   "current": false,
   "price": "Free",
   "commercial": {
    "value": "With conditions",
    "status": "cond"
   },
   "access": "Download"
  },
  {
   "slug": "stable-video-diffusion",
   "name": "Stable Video Diffusion (SVD-XT)",
   "why": "Pick Stable Video Diffusion when you only need to animate one still picture into a short 25-frame clip.",
   "current": false,
   "price": "Free",
   "commercial": {
    "value": "With conditions",
    "status": "cond"
   },
   "access": "Download"
  },
  {
   "slug": "hypit",
   "name": "Hypit",
   "why": "Open-source video toolkit for AI agents that clones reference videos into variants, plus a hosted multi-model video API.",
   "current": false,
   "price": "Free tier",
   "commercial": {
    "value": "With conditions",
    "status": "cond"
   },
   "access": "API, npm"
  },
  {
   "slug": "huobao-drama",
   "name": "Huobao Drama",
   "why": "Open-source platform that automates AI short-drama production from script and characters to storyboard and video compositing.",
   "current": false,
   "price": "Free",
   "commercial": {
    "value": "Non-commercial",
    "status": "no"
   },
   "access": "Download"
  }
 ],
 "faq": [
  {
   "q": "Can I use CogVideoX in a paid product?",
   "a": "The 2B model and the code, yes, under Apache 2.0. The 5B models need a free commercial sign-up, and above one million visits a month you must contact the team."
  },
  {
   "q": "Who owns the videos I make?",
   "a": "The license says the rights to generated content belong to the user, as far as local law allows."
  },
  {
   "q": "Can I try it without my own graphics card?",
   "a": "Yes. There is an online demo on Hugging Face, and the project links Colab notebooks that run on free T4 machines."
  },
  {
   "q": "Why do short prompts give poor clips?",
   "a": "The models were trained on long descriptions, so the team suggests expanding a short idea with a chat AI first."
  }
 ],
 "at_a_glance": {
  "latest_version": "0.42.1",
  "last_release": "2024-11-08",
  "last_commit": "2025-11-04",
  "runtime": "Python 3.10-3.12 per the project, but diffusers 0.42.1 itself needs Python 3.11+, so use 3.11 or 3.12; torch 2.8+ and an NVIDIA GPU (CUDA) for the docs examples.",
  "key_needed": false
 },
 "good_to_know": [
  "You need a strong NVIDIA graphics card or a rented one. One clip takes minutes, not seconds, even on top data-center cards.",
  "Write prompts in English. Other languages should be translated first, for example by a chat AI.",
  "Clips are short and small. The first models give about six seconds at 720 by 480; version 1.5 gives five or ten seconds at a higher size.",
  "The 5B models may not be used for military or illegal purposes, and their license can be updated later.",
  "The project now lives under the zai-org account; old THUDM links forward there. The newest models were released in November 2024."
 ],
 "stack_snippets": [],
 "claims": [
  "The larger model takes about 180 seconds per clip on a single A100 card",
  "The smallest model can run with about 4GB of graphics memory when all memory savings are on"
 ],
 "checked": {
  "license_read": "2026-11-10",
  "overview_checked": "2026-11-10",
  "numbers_updated": "2026-11-19",
  "sources": [
   {
    "label": "github.com/THUDM/CogVideo",
    "url": "https://github.com/THUDM/CogVideo"
   },
   {
    "label": "github.com/zai-org/CogVideo",
    "url": "https://github.com/zai-org/CogVideo"
   },
   {
    "label": "README on GitHub",
    "url": "https://github.com/zai-org/CogVideo/blob/HEAD/README.md"
   },
   {
    "label": "LICENSE on GitHub",
    "url": "https://github.com/zai-org/CogVideo/blob/HEAD/LICENSE"
   },
   {
    "label": "LICENSE on GitHub",
    "url": "https://github.com/zai-org/CogVideo/blob/HEAD/MODEL_LICENSE"
   },
   {
    "label": "github.com/zai-org/CogVideo/blob/HEAD/requirements.txt",
    "url": "https://github.com/zai-org/CogVideo/blob/HEAD/requirements.txt"
   },
   {
    "label": "github.com/zai-org/CogVideo/blob/HEAD/inference/convert_demo.py",
    "url": "https://github.com/zai-org/CogVideo/blob/HEAD/inference/convert_demo.py"
   },
   {
    "label": "huggingface.co",
    "url": "https://huggingface.co/THUDM/CogVideoX-5b/raw/main/README.md"
   },
   {
    "label": "huggingface.co",
    "url": "https://huggingface.co/THUDM/CogVideoX-5b/raw/main/LICENSE"
   },
   {
    "label": "huggingface.co",
    "url": "https://huggingface.co/THUDM/CogVideoX-5b-I2V/raw/main/README.md"
   },
   {
    "label": "huggingface.co",
    "url": "https://huggingface.co/THUDM/CogVideoX-2b/raw/main/README.md"
   },
   {
    "label": "huggingface.co",
    "url": "https://huggingface.co/docs/diffusers/api/pipelines/cogvideox"
   },
   {
    "label": "pypi.org/pypi/diffusers/0.42.1/json",
    "url": "https://pypi.org/pypi/diffusers/0.42.1/json"
   },
   {
    "label": "LICENSE on GitHub",
    "url": "https://github.com/THUDM/CogVideo/blob/HEAD/LICENSE"
   }
  ],
  "unverified": null
 },
 "user_reports": null,
 "url": "https://agentgid.com/assets/cogvideo/",
 "note": "Written by AgentGid from the official pages in checked.sources; quotes are copied from those pages. status: yes | cond | no | unknown | info. Prices and vendor claims are as reported on the checked date. Not legal advice."
}
