{
 "slug": "triposg",
 "name": "TripoSG",
 "author": "VAST AI Research (Tripo)",
 "density": "full",
 "status": "active",
 "status_note": null,
 "alternatives_page": null,
 "one_liner": "1.5B rectified-flow model that turns a single image into a detailed 3D mesh.",
 "category": {
  "slug": "ai-3d",
  "name": "AI 3D model generators"
 },
 "tags": [
  "Open source"
 ],
 "styles": [],
 "logo_url": "https://agentgid.com/media/avatars/vast-ai-research.png",
 "logo_letter": "T",
 "links": {
  "repo": "https://github.com/VAST-AI-Research/TripoSG",
  "website": "https://huggingface.co/VAST-AI/TripoSG",
  "api_docs": null
 },
 "preview": null,
 "verdicts": [
  {
   "key": "price",
   "label": "Price",
   "value": "Free",
   "status": "yes",
   "sub": "No paid plan",
   "quote": null,
   "note": "Open weights, free to download.",
   "source": {
    "label": "huggingface.co/VAST-AI/TripoSG",
    "url": "https://huggingface.co/VAST-AI/TripoSG"
   },
   "n": 1
  },
  {
   "key": "commercial",
   "label": "Commercial use",
   "value": "Yes",
   "status": "yes",
   "sub": "MIT",
   "quote": "to use, copy, modify, merge, publish, distribute, sublicense, and/or sell",
   "note": "TripoSG code and weights are MIT; the RMBG-1.4 background model used by the default script is non-commercial without a BRIA license.",
   "source": {
    "label": "LICENSE on GitHub",
    "url": "https://github.com/VAST-AI-Research/TripoSG/blob/main/LICENSE"
   },
   "n": 2
  },
  {
   "key": "credit",
   "label": "Credit the author",
   "value": "Not needed",
   "status": "yes",
   "sub": "Keep the notice in code copies",
   "quote": "The above copyright notice and this permission notice shall be included in all",
   "note": "No credit on screen. Keep the license notice inside the code you ship.",
   "source": {
    "label": "LICENSE on GitHub",
    "url": "https://github.com/VAST-AI-Research/TripoSG/blob/main/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/VAST-AI-Research/TripoSG",
    "url": "https://github.com/VAST-AI-Research/TripoSG"
   },
   "n": 4
  }
 ],
 "license": {
  "name": "MIT",
  "url": "https://github.com/VAST-AI-Research/TripoSG/blob/main/LICENSE",
  "summary": "The TripoSG code and model are under the MIT license, so paid use is allowed if you keep the license notice with copies. The background cutter the script downloads, RMBG-1.4, is free only for non-commercial use.",
  "pages_read": 1,
  "rules": [
   "Keep the MIT license file with the code or weights you ship.",
   "Needs a CUDA GPU with more than 8 GB of VRAM (model card)."
  ],
  "rows": [
   {
    "label": "Commercial use",
    "value": "Yes",
    "status": "yes",
    "quote": "to use, copy, modify, merge, publish, distribute, sublicense, and/or sell",
    "source": {
     "label": "LICENSE on GitHub",
     "url": "https://github.com/VAST-AI-Research/TripoSG/blob/main/LICENSE"
    },
    "note": null
   },
   {
    "label": "Credit",
    "value": "Not needed",
    "status": "yes",
    "quote": "The above copyright notice and this permission notice shall be included in all",
    "source": {
     "label": "LICENSE on GitHub",
     "url": "https://github.com/VAST-AI-Research/TripoSG/blob/main/LICENSE"
    },
    "note": null
   },
   {
    "label": "Inside your app or game",
    "value": "Can ship inside your app",
    "status": "yes",
    "quote": "in the Software without restriction, including without limitation the rights",
    "source": {
     "label": "LICENSE on GitHub",
     "url": "https://github.com/VAST-AI-Research/TripoSG/blob/main/LICENSE"
    },
    "note": null
   },
   {
    "label": "Pass the files on to your users",
    "value": "With conditions",
    "status": "cond",
    "quote": "to use, copy, modify, merge, publish, distribute, sublicense, and/or sell",
    "source": {
     "label": "LICENSE on GitHub",
     "url": "https://github.com/VAST-AI-Research/TripoSG/blob/main/LICENSE"
    },
    "note": "You may copy and pass on the TripoSG code and weights with the MIT notice."
   },
   {
    "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": "2026-10-25",
  "commercial_use_page": "https://agentgid.com/assets/triposg/commercial-use/"
 },
 "overview": {
  "heading": "TripoSG: a detailed 3D shape from one picture",
  "body": "TripoSG is a free AI model that turns one picture of an object into a 3D model file. It works with photos, cartoons and sketches, and gives a shape without colors. It runs on your own computer and needs an NVIDIA graphics card with at least 8 GB of memory."
 },
 "uses": [
  {
   "title": "A 3D shape from a product photo",
   "text": "Turn one picture of a chair, toy or gadget into a GLB file you can open in Blender or a game engine.",
   "idea": false,
   "image": null
  },
  {
   "title": "Props for games and visual effects",
   "text": "Make 3D stand-ins or base shapes for game props and effects shots from concept art, then paint and tune them in your 3D tool.",
   "idea": false,
   "image": null
  },
  {
   "title": "A lighter model for the web",
   "text": "Ask for fewer triangles so the file stays small enough for a web page or a phone game.",
   "idea": false,
   "image": null
  },
  {
   "title": "A 3D sketch from a doodle and words",
   "text": "Draw a rough doodle, add a short text like \"a cat with wings\", and get a quick 3D draft to try ideas.",
   "idea": false,
   "image": null
  },
  {
   "title": "Shapes for 3D printing drafts",
   "text": "Make a first shape from a drawing of a figurine or part, then fix and close it in a 3D tool before printing.",
   "idea": true,
   "image": null
  },
  {
   "title": "Turnable items on a portfolio page",
   "text": "Show the GLB file on a page with a 3D viewer so visitors can turn the object with the mouse.",
   "idea": true,
   "image": null
  }
 ],
 "start": {
  "prompt": "Set up TripoSG from github.com/VAST-AI-Research/TripoSG in a Python 3.10 virtual environment with the PyTorch build for my CUDA version, give it product.png with a transparent background, and save a GLB with at most 20000 faces to output/product.glb.",
  "install": null,
  "install_note": null,
  "code": [
   {
    "tab": "Step 1",
    "intro": "Get the code from GitHub, make a Python 3.10 environment, install PyTorch for your CUDA version, then the project requirements.",
    "source": "git clone https://github.com/VAST-AI-Research/TripoSG.git\ncd TripoSG\npip install -r requirements.txt"
   },
   {
    "tab": "Step 2",
    "intro": "Run the script on a picture. The model files are downloaded from Hugging Face the first time.",
    "source": "python -m scripts.inference_triposg --image-input assets/example_data/hjswed.png --output-path ./output.glb"
   }
  ]
 },
 "mistakes": [
  {
   "dont": "Installing the default PyTorch from PyPI and running on a machine without an NVIDIA card",
   "fix": "Install the PyTorch build for your CUDA version from download.pytorch.org and run on a CUDA GPU with 8 GB or more",
   "why": "The script puts the models on \"cuda\" and the README asks for the matching CUDA build.",
   "dont_code": false,
   "fix_code": false
  },
  {
   "dont": "Shipping paid work made with the default script without thinking about RMBG-1.4",
   "fix": "Give it pictures with a clean transparent background so the cut-out step is not used, or get a commercial RMBG license",
   "why": "The script downloads and loads RMBG-1.4, which is non-commercial; a picture with a usable transparent layer skips that model.",
   "dont_code": false,
   "fix_code": false
  },
  {
   "dont": "Running the script from inside the scripts folder (python inference_triposg.py)",
   "fix": "Run it from the project root as a module, python -m scripts.inference_triposg",
   "why": "The README runs it as a module from the root so imports and the pretrained_weights/ folder resolve.",
   "dont_code": false,
   "fix_code": false
  },
  {
   "dont": "Putting the raw GLB with hundreds of thousands of faces on a web page",
   "fix": "Pass --faces (for example 5000) to shrink it while it is made",
   "why": "Without --faces (default -1) the mesh is not reduced.",
   "dont_code": false,
   "fix_code": false
  }
 ],
 "agent": {
  "self_serve": false,
  "self_serve_text": null,
  "install": null,
  "repo": "https://github.com/VAST-AI-Research/TripoSG",
  "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": "3D shapes made with TripoSG (VAST AI Research, MIT License)",
  "rules_text": "- TripoSG (AI 3D model generators). 1.5B rectified-flow model that turns a single image into a detailed 3D mesh. Price: free; commercial use OK; no credit needed.\n  Connect: download from https://huggingface.co/VAST-AI/TripoSG (no API).\n  Rule: Keep the MIT license file with the code or weights you ship.\n  Rule: Needs a CUDA GPU with more than 8 GB of VRAM (model card).\n  License: https://github.com/VAST-AI-Research/TripoSG/blob/main/LICENSE (checked 2026-10-25; details: https://agentgid.com/assets/triposg/)",
  "agent_md": "https://agentgid.com/assets/triposg/agent.md",
  "card_json": "https://agentgid.com/assets/triposg/card.json"
 },
 "popularity": {
  "gid_score": 15,
  "stars": 1924,
  "downloads_7d": null,
  "downloads_7d_label": null,
  "downloads_change_4w_pct": null,
  "eco": null,
  "package": null,
  "repo": "VAST-AI-Research/TripoSG",
  "agent_projects": null,
  "agent_projects_total": null,
  "downloads_unconfirmed": false,
  "series": {
   "downloads_weekly": [],
   "stars": [
    [
     "2026-11-07",
     1888
    ],
    [
     "2026-11-08",
     1888
    ],
    [
     "2026-11-09",
     1893
    ],
    [
     "2026-11-10",
     1897
    ],
    [
     "2026-11-11",
     1901
    ],
    [
     "2026-11-12",
     1906
    ],
    [
     "2026-11-13",
     1907
    ],
    [
     "2026-11-14",
     1909
    ],
    [
     "2026-11-15",
     1909
    ],
    [
     "2026-11-16",
     1913
    ],
    [
     "2026-11-17",
     1918
    ],
    [
     "2026-11-18",
     1923
    ],
    [
     "2026-11-19",
     1924
    ]
   ]
  },
  "stars_flat_since": null,
  "updated_at": "2026-11-19"
 },
 "alternatives": [
  {
   "slug": "triposg",
   "name": "TripoSG",
   "why": null,
   "current": true,
   "price": "Free",
   "commercial": {
    "value": "Commercial OK",
    "status": "yes"
   },
   "access": "Download"
  },
  {
   "slug": "triposr",
   "name": "TripoSR",
   "why": "Pick TripoSR, the earlier and faster model from the same makers, when about 6 GB of graphics memory and colored results matter more than fine detail.",
   "current": false,
   "price": "Free",
   "commercial": {
    "value": "Commercial OK",
    "status": "yes"
   },
   "access": "Download"
  },
  {
   "slug": "hunyuan3d-2",
   "name": "Hunyuan3D 2.0",
   "why": "Pick Hunyuan3D 2.0 when you want a second model to paint a texture on the shape.",
   "current": false,
   "price": "Free",
   "commercial": {
    "value": "With conditions",
    "status": "cond"
   },
   "access": "Download"
  },
  {
   "slug": "trellis",
   "name": "TRELLIS",
   "why": "Pick TRELLIS when you need 3D Gaussians or radiance fields besides a mesh from one picture.",
   "current": false,
   "price": "Free",
   "commercial": {
    "value": "With conditions",
    "status": "cond"
   },
   "access": "Download"
  },
  {
   "slug": "shap-e",
   "name": "Shap-E",
   "why": "OpenAI's text-to-3D and image-to-3D model that outputs textured meshes or neural radiance fields.",
   "current": false,
   "price": "Free",
   "commercial": {
    "value": "With conditions",
    "status": "cond"
   },
   "access": "Download"
  },
  {
   "slug": "point-e",
   "name": "Point-E",
   "why": "Official code and models for generating 3D point clouds from text or images, plus a model that turns point clouds into meshes.",
   "current": false,
   "price": "Free",
   "commercial": {
    "value": "With conditions",
    "status": "cond"
   },
   "access": "Download"
  }
 ],
 "faq": [
  {
   "q": "Does it give colors or textures?",
   "a": "No. TripoSG makes the shape only; textures come from another tool or model."
  },
  {
   "q": "Can I run it on a Mac or a laptop without NVIDIA?",
   "a": "The makers list a CUDA graphics card with at least 8 GB of memory as the requirement, and the script is written for such a card."
  },
  {
   "q": "Can I use the 3D models in a paid game?",
   "a": "The TripoSG code and model are MIT licensed. For paid work, cut out the background yourself or get a commercial RMBG license, because RMBG-1.4 is non-commercial."
  },
  {
   "q": "Which file does it save?",
   "a": "A GLB file at the path you give with --output-path (./output.glb by default)."
  },
  {
   "q": "Does it work with drawings, not only photos?",
   "a": "Yes. The makers say it handles photos, cartoons and sketches, and a separate scribble model takes a doodle plus a text prompt."
  }
 ],
 "at_a_glance": {
  "latest_version": "main branch checked 2026-11-12; Hugging Face weights revision 2c1c516d22d58db486a058d98d31bb6177344e06",
  "last_commit": "2025-04-18",
  "runtime": "Python 3.10 (suggested conda env), PyTorch built for your CUDA version, numpy==1.22.3, diffusers, transformers, pymeshlab; an NVIDIA GPU with at least 8 GB VRAM.",
  "key_needed": false
 },
 "good_to_know": [
  "You need an NVIDIA graphics card with at least 8 GB of memory. The script is set up for such a card only.",
  "The result is a shape only, without colors or textures. Paint it in a 3D tool or use a model that adds textures.",
  "The script also downloads RMBG-1.4 to cut the object out of the photo. That model is for non-commercial use unless you buy a license from BRIA.",
  "It was trained on 2 million picked image and shape pairs, as the makers report.",
  "There is no paid service or key in this project; the makers link free online demos on Hugging Face."
 ],
 "stack_snippets": [],
 "claims": [
  "1.5B parameter model working on 2048 latent tokens",
  "Trained on 2 million curated image and shape pairs"
 ],
 "checked": {
  "license_read": "2026-10-25",
  "overview_checked": "2026-11-12",
  "numbers_updated": "2026-11-19",
  "sources": [
   {
    "label": "github.com/VAST-AI-Research/TripoSG",
    "url": "https://github.com/VAST-AI-Research/TripoSG"
   },
   {
    "label": "README on GitHub",
    "url": "https://github.com/VAST-AI-Research/TripoSG/blob/main/README.md"
   },
   {
    "label": "LICENSE on GitHub",
    "url": "https://github.com/VAST-AI-Research/TripoSG/blob/main/LICENSE"
   },
   {
    "label": "github.com/VAST-AI-Research/TripoSG/blob/main/requirements.txt",
    "url": "https://github.com/VAST-AI-Research/TripoSG/blob/main/requirements.txt"
   },
   {
    "label": "github.com/VAST-AI-Research/TripoSG/blob/main/scripts/inference_triposg.py",
    "url": "https://github.com/VAST-AI-Research/TripoSG/blob/main/scripts/inference_triposg.py"
   },
   {
    "label": "github.com/VAST-AI-Research/TripoSG/blob/main/scripts/image_process.py",
    "url": "https://github.com/VAST-AI-Research/TripoSG/blob/main/scripts/image_process.py"
   },
   {
    "label": "huggingface.co/VAST-AI/TripoSG",
    "url": "https://huggingface.co/VAST-AI/TripoSG"
   },
   {
    "label": "huggingface.co",
    "url": "https://huggingface.co/VAST-AI/TripoSG/raw/main/README.md"
   },
   {
    "label": "huggingface.co/api/models/VAST-AI/TripoSG",
    "url": "https://huggingface.co/api/models/VAST-AI/TripoSG"
   },
   {
    "label": "huggingface.co",
    "url": "https://huggingface.co/api/models/VAST-AI/TripoSG-scribble"
   },
   {
    "label": "huggingface.co",
    "url": "https://huggingface.co/briaai/RMBG-1.4/raw/main/README.md"
   }
  ],
  "unverified": null
 },
 "user_reports": null,
 "url": "https://agentgid.com/assets/triposg/",
 "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."
}
