Spaces:

robotflowlabs
/

README

Running

App Files Files Community

README / index.html

ilessio-aiflowlab

Upload folder using huggingface_hub

fc510d1 verified 2 days ago

raw

history blame contribute delete

13 kB

	<!DOCTYPE html>
	<html lang="en">
	<head>
	<meta charset="UTF-8">
	<meta name="viewport" content="width=device-width, initial-scale=1.0">
	<title>RobotFlowLabs — Foundation Models for Real Robots</title>
	<link href="https://fonts.googleapis.com/css2?family=Oswald:wght@300;400;700&family=JetBrains+Mono:wght@400;500;700&display=swap" rel="stylesheet">
	<style>
	:root {
	--orange: #FF3B00;
	--black: #050505;
	--offwhite: #F3F3F3;
	--darkgray: #1A1A1A;
	--midgray: #2A2A2A;
	--green: #00CC66;
	--dim: #888;
	}

	* { margin: 0; padding: 0; box-sizing: border-box; }

	body {
	font-family: 'JetBrains Mono', monospace;
	background: var(--black);
	color: var(--offwhite);
	line-height: 1.6;
	overflow-x: hidden;
	}

	body::before {
	content: '';
	position: fixed;
	top: 0; left: 0; right: 0; bottom: 0;
	background:
	repeating-linear-gradient(0deg, transparent, transparent 49px, rgba(255,59,0,0.015) 50px),
	repeating-linear-gradient(90deg, transparent, transparent 49px, rgba(255,59,0,0.015) 50px);
	pointer-events: none;
	z-index: 0;
	}

	.container {
	max-width: 960px;
	margin: 0 auto;
	padding: 0 24px;
	position: relative;
	z-index: 1;
	}

	/* HERO */
	.hero {
	padding: 80px 0 60px;
	text-align: center;
	border-bottom: 4px solid var(--orange);
	}

	.hero-label {
	font-family: 'JetBrains Mono', monospace;
	font-size: 11px;
	letter-spacing: 4px;
	text-transform: uppercase;
	color: var(--orange);
	margin-bottom: 16px;
	}

	.hero h1 {
	font-family: 'Oswald', sans-serif;
	font-size: 56px;
	font-weight: 700;
	text-transform: uppercase;
	letter-spacing: 3px;
	line-height: 1.1;
	margin-bottom: 20px;
	}

	.hero h1 span {
	color: var(--orange);
	}

	.hero-tagline {
	font-size: 16px;
	color: var(--dim);
	max-width: 600px;
	margin: 0 auto 32px;
	line-height: 1.7;
	}

	.hero-stats {
	display: flex;
	justify-content: center;
	gap: 48px;
	margin-top: 40px;
	}

	.stat {
	text-align: center;
	}

	.stat-number {
	font-family: 'Oswald', sans-serif;
	font-size: 42px;
	font-weight: 700;
	color: var(--orange);
	line-height: 1;
	}

	.stat-label {
	font-size: 11px;
	text-transform: uppercase;
	letter-spacing: 2px;
	color: var(--dim);
	margin-top: 4px;
	}

	/* SECTIONS */
	section {
	padding: 60px 0;
	border-bottom: 1px solid var(--midgray);
	}

	section h2 {
	font-family: 'Oswald', sans-serif;
	font-size: 28px;
	text-transform: uppercase;
	letter-spacing: 2px;
	color: var(--orange);
	margin-bottom: 24px;
	}

	section p {
	color: #CCC;
	margin-bottom: 16px;
	font-size: 14px;
	}

	/* MISSION */
	.mission-text {
	font-size: 18px;
	color: var(--offwhite);
	line-height: 1.8;
	max-width: 700px;
	}

	.mission-text strong {
	color: var(--orange);
	}

	/* GRID */
	.collection-grid {
	display: grid;
	grid-template-columns: 1fr 1fr;
	gap: 16px;
	margin-top: 24px;
	}

	.collection-card {
	background: var(--darkgray);
	border: 1px solid var(--midgray);
	padding: 24px;
	transition: border-color 0.2s;
	text-decoration: none;
	color: inherit;
	display: block;
	}

	.collection-card:hover {
	border-color: var(--orange);
	}

	.collection-card h3 {
	font-family: 'Oswald', sans-serif;
	font-size: 18px;
	text-transform: uppercase;
	letter-spacing: 1px;
	color: var(--offwhite);
	margin-bottom: 8px;
	}

	.collection-card .tag {
	display: inline-block;
	font-size: 10px;
	text-transform: uppercase;
	letter-spacing: 1px;
	color: var(--orange);
	border: 1px solid var(--orange);
	padding: 2px 8px;
	margin-bottom: 12px;
	}

	.collection-card p {
	font-size: 12px;
	color: var(--dim);
	margin: 0;
	}

	.collection-card .models {
	font-size: 11px;
	color: #666;
	margin-top: 12px;
	border-top: 1px solid var(--midgray);
	padding-top: 8px;
	}

	/* PIPELINE */
	.pipeline {
	display: flex;
	align-items: center;
	justify-content: center;
	gap: 8px;
	margin: 32px 0;
	flex-wrap: wrap;
	}

	.pipeline-step {
	background: var(--darkgray);
	border: 1px solid var(--midgray);
	padding: 12px 20px;
	text-align: center;
	font-size: 12px;
	min-width: 120px;
	}

	.pipeline-step .step-name {
	font-family: 'Oswald', sans-serif;
	font-size: 14px;
	text-transform: uppercase;
	color: var(--offwhite);
	}

	.pipeline-step .step-detail {
	color: var(--dim);
	font-size: 10px;
	margin-top: 4px;
	}

	.pipeline-arrow {
	color: var(--orange);
	font-size: 20px;
	font-weight: bold;
	}


	/* FOOTER */
	footer {
	padding: 40px 0;
	text-align: center;
	color: var(--dim);
	font-size: 11px;
	letter-spacing: 1px;
	}

	footer a {
	color: var(--orange);
	text-decoration: none;
	}

	.badge {
	display: inline-block;
	background: var(--orange);
	color: var(--black);
	font-size: 10px;
	font-weight: 700;
	padding: 2px 8px;
	text-transform: uppercase;
	letter-spacing: 1px;
	vertical-align: middle;
	margin-left: 8px;
	}

	@media (max-width: 640px) {
	.hero h1 { font-size: 36px; }
	.hero-stats { gap: 24px; flex-wrap: wrap; }
	.stat-number { font-size: 32px; }
	.collection-grid { grid-template-columns: 1fr; }
	.pipeline { flex-direction: column; }
	.pipeline-arrow { transform: rotate(90deg); }
	}
	</style>
	</head>
	<body>

	<div class="container">

	<!-- HERO -->
	<div class="hero">
	<div class="hero-label">/// RobotFlowLabs</div>
	<h1><span>Foundation Models</span><br>for Real Robots</h1>
	<p class="hero-tagline">
	We optimize vision, language, and action models for real-time edge deployment.
	Every model here is compressed, benchmarked, and ready for production robotics.
	</p>

	<div class="hero-stats">
	<div class="stat">
	<div class="stat-number">16+</div>
	<div class="stat-label">Models</div>
	</div>
	<div class="stat">
	<div class="stat-number">58</div>
	<div class="stat-label">ANIMA Modules</div>
	</div>
	<div class="stat">
	<div class="stat-number">4.5x</div>
	<div class="stat-label">Max Compression</div>
	</div>
	<div class="stat">
	<div class="stat-number">ROS2</div>
	<div class="stat-label">Native</div>
	</div>
	</div>
	</div>

	<!-- MISSION -->
	<section>
	<h2>Mission</h2>
	<p class="mission-text">
	The robotics community deserves <strong>production-ready models</strong>, not just research checkpoints.
	We take the best open foundation models — CLIP, SAM2, DINOv2, Qwen, Depth Anything — and make them
	<strong>actually deployable</strong> on the hardware robots use: Jetson Orin, industrial PCs, edge GPUs.
	</p>
	<p class="mission-text" style="margin-top: 16px;">
	Every model is quantized (INT4/INT8), exported (ONNX/SafeTensors/TorchScript), and benchmarked on
	real hardware. No guesswork, no "should work in theory" — <strong>measured performance on real silicon</strong>.
	</p>
	</section>

	<!-- COLLECTIONS -->
	<section>
	<h2>Model Collections</h2>
	<p>Organized by capability for the ANIMA robotics stack.</p>

	<div class="collection-grid">
	<a class="collection-card" href="https://huggingface.co/collections/robotflowlabs/anima-vision-69bc77ca7ce15b06bbdd21bd">
	<div class="tag">Perception</div>
	<h3>ANIMA Vision</h3>
	<p>Segmentation, features, depth estimation, and visual grounding for robotic scene understanding.</p>
	<div class="models">SAM2 · DINOv2 · CLIP · SigLIP · Depth Anything</div>
	</a>
	<a class="collection-card" href="https://huggingface.co/collections/robotflowlabs/anima-language-69bc77ca29dccc3f68f8c7fd">
	<div class="tag">Reasoning</div>
	<h3>ANIMA Language</h3>
	<p>INT4 quantized language models for instruction following, planning, and robotic reasoning.</p>
	<div class="models">Qwen2.5-7B · SmolLM2-1.7B</div>
	</a>
	<a class="collection-card" href="https://huggingface.co/collections/robotflowlabs/anima-vlm-69bc77ca53ae84ac21b0f012">
	<div class="tag">Understanding</div>
	<h3>ANIMA VLM</h3>
	<p>Vision-language models for visual QA, scene description, and grounding language to observations.</p>
	<div class="models">Qwen2.5-VL-7B · Qwen2.5-VL-3B</div>
	</a>
	<a class="collection-card" href="https://huggingface.co/collections/robotflowlabs/anima-vla-69bc77cbf1b8aa40002920bb">
	<div class="tag">Action</div>
	<h3>ANIMA VLA</h3>
	<p>Vision-Language-Action models for end-to-end robotic control and manipulation.</p>
	<div class="models">SmolVLA · RDT2-FM · FORGE Students</div>
	</a>
	</div>
	</section>

	<!-- FORGE PIPELINE -->
	<section>
	<h2>FORGE — Compression Pipeline</h2>
	<p>Our 4-stage pipeline takes any 7B+ VLA model down to <2GB for real-time edge deployment.</p>

	<div class="pipeline">
	<div class="pipeline-step">
	<div class="step-name">Teacher Labels</div>
	<div class="step-detail">5 teachers, multi-GPU</div>
	</div>
	<div class="pipeline-arrow">▶</div>
	<div class="pipeline-step">
	<div class="step-name">Distillation</div>
	<div class="step-detail">KD + curriculum</div>
	</div>
	<div class="pipeline-arrow">▶</div>
	<div class="pipeline-step">
	<div class="step-name">Compression</div>
	<div class="step-detail">Prune + quantize</div>
	</div>
	<div class="pipeline-arrow">▶</div>
	<div class="pipeline-step">
	<div class="step-name">Export</div>
	<div class="step-detail">ONNX / TRT / MLX</div>
	</div>
	</div>

	<p style="color: var(--dim); font-size: 12px; text-align: center;">
	Automated hyperparameter optimization via Optuna · 400+ trials across 4 GPUs · W&B experiment tracking
	</p>
	</section>


	</div>

	<!-- FOOTER -->
	<footer>
	<div class="container">
	<p>
	<a href="https://huggingface.co/robotflowlabs">RobotFlowLabs</a>
	· Building the future of robotic intelligence
	· 2026
	</p>
	<p style="margin-top: 8px; color: #444;">
	Models compressed with FORGE · Benchmarked on NVIDIA L4 · All weights open
	</p>
	</div>
	</footer>

	</body>
	</html>