Open-source platform for 3D AI agents. Turn text or a photo into a rigged, animated GLB avatar, give it an LLM brain, memory and a wallet, and embed it anywhere with one web component. Remote MCP server, x402 payments, Three.js, browser-native.
JavaScript 3d-generationai-agentsanimationavatardigital-humangenerative-aiglbgltfimage-to-3dllmmcpmcp-servermodel-context-protocolsolanatext-to-3dthreejsweb-componentswebglwebxrx402
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8" />
<meta name="viewport" content="width=device-width, initial-scale=1.0" />
<title data-i18n="features_forge.meta_title">Forge — Turn Text or a Photo into a 3D Model · three.ws</title>
<meta
name="description"
content="Type a description or upload photos, and get a downloadable 3D model. Forge runs multiple self-hosted engines to produce a textured GLB you can inspect in AR, embed anywhere, and attach to your AI agent."
data-i18n-attr="content:features_forge.meta_description"/>
<meta name="robots" content="index, follow" />
<link rel="canonical" href="https://three.ws/features/forge" />
<meta property="og:type" content="website" />
<meta property="og:site_name" content="three.ws" />
<meta property="og:title" content="Forge — Turn Text or a Photo into a 3D Model · three.ws" data-i18n-attr="content:features_forge.meta_og_title"/>
<meta
property="og:description"
content="Type a description or upload photos, and get a downloadable 3D model in one click."
data-i18n-attr="content:features_forge.meta_og_description"/>
<meta property="og:url" content="https://three.ws/features/forge" />
<meta
property="og:image"
content="https://three.ws/api/feature-og?feature=forge&title=Type+a+description.+Get+a+3D+model.&sub=Forge+turns+text+into+a+downloadable+GLB+in+seconds."
/>
<meta property="og:image:width" content="1200" />
<meta property="og:image:height" content="630" />
<meta name="twitter:card" content="summary_large_image" />
<meta name="twitter:title" content="Forge — Turn Text or a Photo into a 3D Model · three.ws" data-i18n-attr="content:features_forge.meta_twitter_title"/>
<meta
name="twitter:description"
content="Type a description or upload photos, and get a downloadable 3D model in one click."
data-i18n-attr="content:features_forge.meta_twitter_description"/>
<meta
name="twitter:image"
content="https://three.ws/api/feature-og?feature=forge&title=Type+a+description.+Get+a+3D+model.&sub=Forge+turns+text+into+a+downloadable+GLB+in+seconds."
/>
<link rel="shortcut icon" href="/favicon.ico" />
<link rel="apple-touch-icon" href="/pwa-192x192.png" />
<link rel="stylesheet" href="/style.css" />
<link rel="stylesheet" href="/nav.css" />
<link rel="stylesheet" href="/features-landing.css" />
<link rel="stylesheet" href="/fonts/fonts.css" />
<script
type="module"
src="https://cdn.jsdelivr.net/npm/@google/[email protected]/dist/model-viewer.min.js"
integrity="sha384-wjsPzEO+lXOf7d9csIDfAQpPt2OKuUU3m0K9qsdKIuzX4N8czSF6hdh23eAbZa3p"
crossorigin="anonymous"
></script>
<script type="application/ld+json">
{
"@context": "https://schema.org",
"@graph": [
{
"@type": "SoftwareApplication",
"name": "Forge — Text & Image to 3D",
"url": "https://three.ws/forge",
"applicationCategory": "DesignApplication",
"description": "Generate textured 3D models from text descriptions using Flux image generation and TRELLIS 3D reconstruction.",
"operatingSystem": "Web",
"offers": { "@type": "Offer", "price": "0", "priceCurrency": "USD" }
},
{
"@type": "FAQPage",
"mainEntity": [
{
"@type": "Question",
"name": "How long does Forge take to generate a model?",
"acceptedAnswer": {
"@type": "Answer",
"text": "Most models generate in 30–90 seconds. Flux creates the texture image first, then TRELLIS reconstructs the 3D mesh from it."
}
},
{
"@type": "Question",
"name": "What file format does Forge output?",
"acceptedAnswer": {
"@type": "Answer",
"text": "Forge outputs a GLB (binary glTF) file — the industry-standard 3D format readable by Blender, Unity, Unreal, and every major 3D viewer."
}
},
{
"@type": "Question",
"name": "Can I use the model commercially?",
"acceptedAnswer": {
"@type": "Answer",
"text": "Yes. Models you generate with Forge are yours to use, modify, and ship in any project including commercial work."
}
}
]
}
]
}
</script>
<script src="/model-viewer-meshopt.js"></script>
</head>
<body class="fl-page">
<header><div id="nav-container"></div></header>
<script src="/nav.js"></script>
<main class="fl-main" id="main-content">
<nav class="fl-breadcrumb" aria-label="Breadcrumb">
<a href="/" data-i18n="features_forge.three_ws">three.ws</a>
<span class="fl-breadcrumb-sep" aria-hidden="true">/</span>
<a href="/features" data-i18n="features_forge.features">Features</a>
<span class="fl-breadcrumb-sep" aria-hidden="true">/</span>
<span class="fl-breadcrumb-current" data-i18n="features_forge.forge">Forge</span>
</nav>
<!-- Hero -->
<section class="fl-hero" aria-labelledby="forge-hero-h1">
<div class="fl-hero-inner">
<div class="fl-hero-copy">
<span class="fl-eyebrow"
data-i18n-html="features_forge.text_image_to_3d"><span class="fl-eyebrow-dot" aria-hidden="true"></span>Text & Image to 3D</span
>
<h1 id="forge-hero-h1" data-i18n-html="features_forge.type_a_description_get_a_3d_model">Type a description.<br />Get a 3D model.</h1>
<p class="fl-hero-sub" data-i18n="features_forge.forge_runs_flux_image_generation_and_trellis">
Forge runs Flux image generation and TRELLIS 3D reconstruction
back-to-back. Describe any object and walk away with a textured GLB you
can orbit in AR, embed anywhere, or attach to your AI agent.
</p>
<div class="fl-hero-actions">
<a href="/forge" class="fl-btn-primary" data-i18n="features_forge.open_forge">Open Forge →</a>
<a href="/features" class="fl-btn-ghost" data-i18n="features_forge.all_features">All features</a>
</div>
</div>
<div
class="fl-hero-preview"
aria-label="3D model preview — robot expressive sample"
>
<model-viewer
src="/animations/robotexpressive.glb"
alt="Sample 3D model generated by Forge — robot character"
auto-rotate
auto-rotate-delay="500"
rotation-per-second="12deg"
camera-controls
shadow-intensity="0.4"
environment-image="neutral"
style="background: transparent; width: 100%; height: 100%"
></model-viewer>
<span class="fl-preview-label" data-i18n="features_forge.sample_glb_drag_to_orbit">Sample GLB · drag to orbit</span>
</div>
</div>
</section>
<!-- How it works -->
<section class="fl-how fl-section" aria-labelledby="forge-how-h2">
<p class="fl-section-label" data-i18n="features_forge.how_it_works">How it works</p>
<h2 id="forge-how-h2" data-i18n="features_forge.three_steps_from_prompt_to_model">Three steps from prompt to model</h2>
<div class="fl-steps" role="list">
<article class="fl-step" role="listitem">
<span class="fl-step-num">01</span>
<h3 data-i18n="features_forge.write_your_prompt">Write your prompt</h3>
<p data-i18n="features_forge.describe_what_you_want_a_ceramic_coffee">
Describe what you want — "a ceramic coffee mug with blue glaze," "a
rusted iron sword," anything. Be specific about material and style for
best results.
</p>
</article>
<article class="fl-step" role="listitem">
<span class="fl-step-num">02</span>
<h3 data-i18n="features_forge.flux_trellis_generate_it">Flux + TRELLIS generate it</h3>
<p data-i18n="features_forge.flux_renders_a_detailed_texture_image_from">
Flux renders a detailed texture image from your text. TRELLIS lifts that
image into a full 3D mesh with UV-mapped textures, all in one pipeline
run.
</p>
</article>
<article class="fl-step" role="listitem">
<span class="fl-step-num">03</span>
<h3 data-i18n="features_forge.download_embed_or_attach">Download, embed, or attach</h3>
<p data-i18n="features_forge.inspect_the_result_in_the_in_browser">
Inspect the result in the in-browser viewer, download the GLB, share a
link, or pin it to your AI agent as its 3D body — all from the same
page.
</p>
</article>
</div>
</section>
<!-- Highlight cards -->
<section class="fl-highlights fl-section" aria-labelledby="forge-highlights-h2">
<h2 id="forge-highlights-h2" data-i18n="features_forge.what_you_can_do_with_it">What you can do with it</h2>
<div class="fl-cards">
<div class="fl-card">
<span class="fl-card-icon" aria-hidden="true">⬡</span>
<h3 data-i18n="features_forge.ar_inspection">AR inspection</h3>
<p data-i18n="features_forge.every_generated_model_opens_in_webxr_ar">
Every generated model opens in WebXR AR mode with one tap — place it on
your real desk or floor and walk around it.
</p>
</div>
<div class="fl-card">
<span class="fl-card-icon" aria-hidden="true">⬡</span>
<h3 data-i18n="features_forge.agent_avatar">Agent avatar</h3>
<p data-i18n="features_forge.attach_the_glb_to_any_ai_agent">
Attach the GLB to any AI agent on three.ws so the agent gets a custom 3D
body instead of a default avatar.
</p>
</div>
<div class="fl-card">
<span class="fl-card-icon" aria-hidden="true">⬡</span>
<h3 data-i18n="features_forge.shareable_link">Shareable link</h3>
<p data-i18n="features_forge.each_generated_model_gets_a_permanent_url">
Each generated model gets a permanent URL you can share — it loads in
any browser without plugins or app installs.
</p>
</div>
<div class="fl-card">
<span class="fl-card-icon" aria-hidden="true">⬡</span>
<h3 data-i18n="features_forge.data_flywheel">Data flywheel</h3>
<p data-i18n="features_forge.all_forge_generations_feed_the_three_ws">
All forge generations feed the three.ws training flywheel — improving
model quality for every creator over time.
</p>
</div>
</div>
</section>
<!-- FAQ -->
<section class="fl-faq fl-section" aria-labelledby="forge-faq-h2">
<h2 id="forge-faq-h2" data-i18n="features_forge.frequently_asked_questions">Frequently asked questions</h2>
<div class="fl-faq-list">
<div class="fl-faq-item">
<button class="fl-faq-q" aria-expanded="false" aria-controls="forge-faq-1">
How long does generation take?
<svg
class="fl-faq-chevron"
viewBox="0 0 16 16"
fill="none"
aria-hidden="true"
>
<path
d="M4 6l4 4 4-4"
stroke="currentColor"
stroke-width="1.5"
stroke-linecap="round"
/>
</svg>
</button>
<div class="fl-faq-a" id="forge-faq-1" role="region">
Most models complete in 30–90 seconds. Flux renders the texture image
first, then TRELLIS reconstructs the mesh — the progress bar shows both
stages.
</div>
</div>
<div class="fl-faq-item">
<button class="fl-faq-q" aria-expanded="false" aria-controls="forge-faq-2">
What file format does Forge export?
<svg
class="fl-faq-chevron"
viewBox="0 0 16 16"
fill="none"
aria-hidden="true"
>
<path
d="M4 6l4 4 4-4"
stroke="currentColor"
stroke-width="1.5"
stroke-linecap="round"
/>
</svg>
</button>
<div class="fl-faq-a" id="forge-faq-2" role="region">
GLB (binary glTF 2.0) — the industry-standard format supported by
Blender, Unity, Unreal Engine, and every major 3D viewer including Apple
Quick Look.
</div>
</div>
<div class="fl-faq-item">
<button class="fl-faq-q" aria-expanded="false" aria-controls="forge-faq-3">
Can I use generated models commercially?
<svg
class="fl-faq-chevron"
viewBox="0 0 16 16"
fill="none"
aria-hidden="true"
>
<path
d="M4 6l4 4 4-4"
stroke="currentColor"
stroke-width="1.5"
stroke-linecap="round"
/>
</svg>
</button>
<div class="fl-faq-a" id="forge-faq-3" role="region">
Yes. Models you generate are yours to use, modify, and ship in any
project, including commercial products.
</div>
</div>
<div class="fl-faq-item">
<button class="fl-faq-q" aria-expanded="false" aria-controls="forge-faq-4">
What prompts work best?
<svg
class="fl-faq-chevron"
viewBox="0 0 16 16"
fill="none"
aria-hidden="true"
>
<path
d="M4 6l4 4 4-4"
stroke="currentColor"
stroke-width="1.5"
stroke-linecap="round"
/>
</svg>
</button>
<div class="fl-faq-a" id="forge-faq-4" role="region">
Isolated objects with clear material descriptions generate the cleanest
meshes. "A ceramic mug with blue glaze" outperforms "a mug." Avoid
scenes with multiple objects or complex backgrounds — those compress
poorly into a single mesh.
</div>
</div>
<div class="fl-faq-item">
<button class="fl-faq-q" aria-expanded="false" aria-controls="forge-faq-5">
Is this different from the selfie scanner?
<svg
class="fl-faq-chevron"
viewBox="0 0 16 16"
fill="none"
aria-hidden="true"
>
<path
d="M4 6l4 4 4-4"
stroke="currentColor"
stroke-width="1.5"
stroke-linecap="round"
/>
</svg>
</button>
<div class="fl-faq-a" id="forge-faq-5" role="region">
Yes.
<a href="/features/scan" style="color: var(--ink)" data-i18n="features_forge.scan">Scan</a> reconstructs
a 3D avatar from a real photo of your face. Forge generates any object
from a text description — they serve different use cases.
</div>
</div>
</div>
</section>
<!-- Final CTA -->
<section class="fl-cta-band" aria-labelledby="forge-cta-h2">
<div class="fl-cta-inner">
<h2 id="forge-cta-h2" data-i18n="features_forge.start_generating_in_30_seconds">Start generating in 30 seconds</h2>
<p data-i18n="features_forge.no_account_required_for_your_first_model">
No account required for your first model. Type a description, watch it
appear.
</p>
<div class="fl-cta-actions">
<a href="/forge" class="fl-btn-primary" data-i18n="features_forge.open_forge_2">Open Forge →</a>
<a href="/features/scan" class="fl-btn-ghost" data-i18n="features_forge.try_scan_instead">Try Scan instead</a>
</div>
</div>
</section>
</main>
<script>
document.querySelectorAll('.fl-faq-q').forEach((btn) => {
btn.addEventListener('click', () => {
const expanded = btn.getAttribute('aria-expanded') === 'true';
btn.setAttribute('aria-expanded', String(!expanded));
const panel = document.getElementById(btn.getAttribute('aria-controls'));
if (panel) panel.classList.toggle('open', !expanded);
});
});
</script>
<div id="footer-container"></div>
<script src="/footer.js"></script>
<!-- Runtime i18n: detects locale, swaps annotated copy, mounts <lang-switcher>. -->
<script type="module" src="/i18n.js"></script>
</body>
</html>