Spaces:
Running
Running
Update index.html
Browse files- index.html +293 -184
index.html
CHANGED
|
@@ -1,199 +1,308 @@
|
|
| 1 |
-
<meta charset="utf-8">
|
| 2 |
<style>
|
| 3 |
-
|
| 4 |
-
|
| 5 |
-
-
|
| 6 |
-
|
| 7 |
-
|
| 8 |
-
|
| 9 |
-
|
| 10 |
-
|
| 11 |
-
|
| 12 |
-
|
| 13 |
-
|
| 14 |
-
|
| 15 |
-
|
| 16 |
-
|
| 17 |
-
|
| 18 |
-
|
| 19 |
-
|
| 20 |
-
|
| 21 |
-
|
| 22 |
-
-
|
| 23 |
-
|
| 24 |
-
|
| 25 |
-
-
|
| 26 |
-
|
| 27 |
-
|
| 28 |
-
|
| 29 |
-
|
| 30 |
-
.
|
| 31 |
-
|
| 32 |
-
|
| 33 |
-
|
| 34 |
-
|
| 35 |
-
.
|
| 36 |
-
|
| 37 |
-
|
| 38 |
-
|
| 39 |
-
|
| 40 |
-
|
| 41 |
-
|
| 42 |
-
|
| 43 |
-
.
|
| 44 |
-
|
| 45 |
-
|
| 46 |
-
|
| 47 |
-
|
| 48 |
-
|
| 49 |
-
|
| 50 |
-
|
| 51 |
-
|
| 52 |
-
|
| 53 |
-
|
| 54 |
-
|
| 55 |
-
|
| 56 |
-
|
| 57 |
-
|
| 58 |
-
|
| 59 |
-
|
| 60 |
-
|
| 61 |
-
.
|
| 62 |
-
|
| 63 |
-
|
| 64 |
-
|
| 65 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 66 |
</style>
|
| 67 |
|
| 68 |
-
<div class="
|
| 69 |
-
<div class="
|
| 70 |
-
<
|
| 71 |
-
|
| 72 |
-
<rect width="60" height="40" fill="#012169"/>
|
| 73 |
-
<path d="M0,0 L60,40 M60,0 L0,40" stroke="#fff" stroke-width="8"/>
|
| 74 |
-
<path d="M0,0 L60,40 M60,0 L0,40" stroke="#C8102E" stroke-width="5"/>
|
| 75 |
-
<path d="M30,0 V40 M0,20 H60" stroke="#fff" stroke-width="13"/>
|
| 76 |
-
<path d="M30,0 V40 M0,20 H60" stroke="#C8102E" stroke-width="8"/>
|
| 77 |
-
</svg>
|
| 78 |
-
EN
|
| 79 |
-
</button>
|
| 80 |
-
<button class="lang-btn" id="btn-fr" onclick="setLang('fr')">
|
| 81 |
-
<svg viewBox="0 0 60 40" xmlns="http://www.w3.org/2000/svg">
|
| 82 |
-
<rect width="20" height="40" fill="#002395"/>
|
| 83 |
-
<rect x="20" width="20" height="40" fill="#fff"/>
|
| 84 |
-
<rect x="40" width="20" height="40" fill="#ED2939"/>
|
| 85 |
-
</svg>
|
| 86 |
-
FR
|
| 87 |
-
</button>
|
| 88 |
</div>
|
| 89 |
|
| 90 |
-
<
|
| 91 |
-
<div class="
|
| 92 |
-
<
|
| 93 |
-
<
|
| 94 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 95 |
|
| 96 |
-
<
|
| 97 |
-
<div class="
|
| 98 |
-
<div class="
|
| 99 |
-
|
| 100 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 101 |
</div>
|
| 102 |
-
<div class="card-title" id="open-title"></div>
|
| 103 |
-
<div class="card-text" id="open-text"></div>
|
| 104 |
-
<hr class="card-divider" style="color:#243907">
|
| 105 |
-
<div class="card-meta" id="open-meta"></div>
|
| 106 |
</div>
|
| 107 |
|
| 108 |
-
<div class="
|
| 109 |
-
<div class="
|
| 110 |
-
|
| 111 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 112 |
</div>
|
| 113 |
-
<div class="card-title" id="prop-title"></div>
|
| 114 |
-
<div class="card-text" id="prop-text"></div>
|
| 115 |
-
<hr class="card-divider" style="color:#B83400">
|
| 116 |
-
<div class="card-meta" id="prop-meta"></div>
|
| 117 |
</div>
|
| 118 |
-
</
|
| 119 |
|
| 120 |
-
<
|
| 121 |
-
<div class="
|
| 122 |
-
|
| 123 |
-
|
|
|
|
| 124 |
<div>
|
| 125 |
-
<
|
| 126 |
-
<
|
| 127 |
-
|
| 128 |
-
|
| 129 |
-
|
| 130 |
-
|
|
|
|
|
|
|
|
|
|
| 131 |
</div>
|
| 132 |
</div>
|
| 133 |
-
|
| 134 |
-
<
|
| 135 |
-
|
| 136 |
-
|
| 137 |
-
|
| 138 |
-
|
| 139 |
-
|
| 140 |
-
|
| 141 |
-
|
| 142 |
-
|
| 143 |
-
|
| 144 |
-
|
| 145 |
-
|
| 146 |
-
|
| 147 |
-
|
| 148 |
-
|
| 149 |
-
|
| 150 |
-
|
| 151 |
-
|
| 152 |
-
releaseDesc: "Une technique clé pour réduire efficacement la taille des modèles de langage tout en préservant leurs performances, les rendant plus faciles à déployer. Nous vous invitons à consulter l'article de blog rédigé sur le sujet pour connaître tous les avantages de cette technique, et à explorer le Space pour parcourir les plus de 5000 modèles basés sur cette technique que nous proposons.",
|
| 153 |
-
releaseBlogLabel: 'Article de blog',
|
| 154 |
-
releaseAppLabel: 'Space HF',
|
| 155 |
-
},
|
| 156 |
-
en: {
|
| 157 |
-
missionLabel: '🎯 Our mission',
|
| 158 |
-
missionTitle: 'Building efficient AI models',
|
| 159 |
-
missionText: "At AlphaEdge, we aim to offer AI models that can run on any type of hardware in a flexible way, via API or Edge, on GPU as well as CPU. Our goal is to deliver high performance on complex tasks while significantly reducing latency, memory consumption, and inference costs. We expose this vision through two channels: our open models and our proprietary models.",
|
| 160 |
-
badgeOpen: 'Open source',
|
| 161 |
-
openTitle: 'Open Models',
|
| 162 |
-
openText: "The open source models available on this Hugging Face organization are based on existing ones, published under permissive licenses, for which we propose improvements. The goal here is to showcase our expertise, particularly in state-of-the-art compression techniques, applied to model classes well known to the community.",
|
| 163 |
-
openMeta: '<a href="https://huggingface.co/AlphaEdge-AI/models" target="_blank" style="color:inherit">Available models</a> · <a href="https://huggingface.co/AlphaEdge-AI/datasets" target="_blank" style="color:inherit">Available datasets</a>',
|
| 164 |
-
badgeProp: 'Proprietary',
|
| 165 |
-
propTitle: 'Commercial Models',
|
| 166 |
-
propText: 'Our proprietary models are built on a new architecture that we call ELM (Efficient Language Models). They offer optimized performance for professional use cases requiring full sovereignty and real-time usage. Resource-efficient, they are available through our API or can be deployed on-premises on your hardware at your company.',
|
| 167 |
-
propMeta: 'Our <a href="https://api-docs.alphaedge-ai.com/" target="_blank" style="color:inherit">API</a> · Our <a href="https://www.alphaedge-ai.com/" target="_blank" style="color:inherit">website</a>',
|
| 168 |
-
releasesLabel: '🗃️ Our main open-source releases',
|
| 169 |
-
releaseName: 'Trimming',
|
| 170 |
-
releaseDesc: "A key technique for effectively reducing the size of language models while preserving their performance, making them easier to deploy. We invite you to read the blog post written on the subject to learn about all the benefits of this technique, and to explore the Space to browse the more than 5,000 models based on this technique that we propose.",
|
| 171 |
-
releaseBlogLabel: 'Blog post',
|
| 172 |
-
releaseAppLabel: 'HF Space',
|
| 173 |
-
}
|
| 174 |
-
};
|
| 175 |
-
|
| 176 |
-
function setLang(lang) {
|
| 177 |
-
const c = T[lang];
|
| 178 |
-
document.getElementById('mission-label').textContent = c.missionLabel;
|
| 179 |
-
document.getElementById('mission-title').textContent = c.missionTitle;
|
| 180 |
-
document.getElementById('mission-text').textContent = c.missionText;
|
| 181 |
-
document.getElementById('badge-open').textContent = c.badgeOpen;
|
| 182 |
-
document.getElementById('open-title').textContent = c.openTitle;
|
| 183 |
-
document.getElementById('open-text').textContent = c.openText;
|
| 184 |
-
document.getElementById('open-meta').innerHTML = c.openMeta;
|
| 185 |
-
document.getElementById('badge-prop').textContent = c.badgeProp;
|
| 186 |
-
document.getElementById('prop-title').textContent = c.propTitle;
|
| 187 |
-
document.getElementById('prop-text').textContent = c.propText;
|
| 188 |
-
document.getElementById('prop-meta').innerHTML = c.propMeta;
|
| 189 |
-
document.getElementById('releases-label').textContent = c.releasesLabel;
|
| 190 |
-
document.getElementById('release-name').textContent = c.releaseName;
|
| 191 |
-
document.getElementById('release-desc').textContent = c.releaseDesc;
|
| 192 |
-
document.getElementById('release-blog-label').textContent = c.releaseBlogLabel;
|
| 193 |
-
document.getElementById('release-app-label').textContent = c.releaseAppLabel;
|
| 194 |
-
document.getElementById('btn-en').classList.toggle('active', lang === 'en');
|
| 195 |
-
document.getElementById('btn-fr').classList.toggle('active', lang === 'fr');
|
| 196 |
-
}
|
| 197 |
-
|
| 198 |
-
setLang('en');
|
| 199 |
-
</script>
|
|
|
|
|
|
|
| 1 |
<style>
|
| 2 |
+
.ae-root {
|
| 3 |
+
font-family: ui-sans-serif, system-ui, -apple-system, BlinkMacSystemFont, "Segoe UI", sans-serif;
|
| 4 |
+
max-width: 720px;
|
| 5 |
+
margin: 0 auto;
|
| 6 |
+
padding: 24px 12px;
|
| 7 |
+
color: #161718;
|
| 8 |
+
}
|
| 9 |
+
|
| 10 |
+
.ae-lang {
|
| 11 |
+
display: inline-flex;
|
| 12 |
+
border: 1.5px solid #01113b;
|
| 13 |
+
border-radius: 999px;
|
| 14 |
+
overflow: hidden;
|
| 15 |
+
margin-bottom: 24px;
|
| 16 |
+
background: #ffffff;
|
| 17 |
+
}
|
| 18 |
+
|
| 19 |
+
.ae-lang span {
|
| 20 |
+
display: inline-flex;
|
| 21 |
+
align-items: center;
|
| 22 |
+
gap: 7px;
|
| 23 |
+
padding: 6px 16px;
|
| 24 |
+
font-size: 12px;
|
| 25 |
+
font-weight: 600;
|
| 26 |
+
color: #01113b;
|
| 27 |
+
}
|
| 28 |
+
|
| 29 |
+
.ae-lang .active {
|
| 30 |
+
background: #01113b;
|
| 31 |
+
color: #d9e7ff;
|
| 32 |
+
}
|
| 33 |
+
|
| 34 |
+
.ae-mission {
|
| 35 |
+
background: #d9e7ff;
|
| 36 |
+
border: 1.5px solid #01113b;
|
| 37 |
+
border-radius: 14px;
|
| 38 |
+
padding: 22px 26px;
|
| 39 |
+
margin-bottom: 22px;
|
| 40 |
+
}
|
| 41 |
+
|
| 42 |
+
.ae-badge {
|
| 43 |
+
display: inline-flex;
|
| 44 |
+
align-items: center;
|
| 45 |
+
gap: 6px;
|
| 46 |
+
border-radius: 999px;
|
| 47 |
+
padding: 4px 13px;
|
| 48 |
+
margin-bottom: 12px;
|
| 49 |
+
font-size: 11px;
|
| 50 |
+
font-weight: 700;
|
| 51 |
+
letter-spacing: .08em;
|
| 52 |
+
text-transform: uppercase;
|
| 53 |
+
}
|
| 54 |
+
|
| 55 |
+
.ae-badge-blue {
|
| 56 |
+
background: #01113b;
|
| 57 |
+
color: #d9e7ff;
|
| 58 |
+
}
|
| 59 |
+
|
| 60 |
+
.ae-title {
|
| 61 |
+
margin: 0 0 10px 0;
|
| 62 |
+
color: #01113b;
|
| 63 |
+
font-size: 22px;
|
| 64 |
+
font-weight: 600;
|
| 65 |
+
}
|
| 66 |
+
|
| 67 |
+
.ae-text {
|
| 68 |
+
margin: 0;
|
| 69 |
+
font-size: 15px;
|
| 70 |
+
line-height: 1.65;
|
| 71 |
+
}
|
| 72 |
+
|
| 73 |
+
.ae-grid {
|
| 74 |
+
display: grid;
|
| 75 |
+
grid-template-columns: repeat(2, minmax(0, 1fr));
|
| 76 |
+
gap: 18px;
|
| 77 |
+
margin-bottom: 22px;
|
| 78 |
+
}
|
| 79 |
+
|
| 80 |
+
.ae-card {
|
| 81 |
+
border-radius: 14px;
|
| 82 |
+
padding: 22px 24px;
|
| 83 |
+
border: 1.5px solid;
|
| 84 |
+
}
|
| 85 |
+
|
| 86 |
+
.ae-open {
|
| 87 |
+
background: #e7ffde;
|
| 88 |
+
border-color: #243907;
|
| 89 |
+
}
|
| 90 |
+
|
| 91 |
+
.ae-prop {
|
| 92 |
+
background: #ffb99d;
|
| 93 |
+
border-color: #b83400;
|
| 94 |
+
}
|
| 95 |
+
|
| 96 |
+
.ae-badge-open {
|
| 97 |
+
background: #243907;
|
| 98 |
+
color: #e7ffde;
|
| 99 |
+
}
|
| 100 |
+
|
| 101 |
+
.ae-badge-prop {
|
| 102 |
+
background: #b83400;
|
| 103 |
+
color: #ffb99d;
|
| 104 |
+
}
|
| 105 |
+
|
| 106 |
+
.ae-card h3 {
|
| 107 |
+
margin: 0 0 8px 0;
|
| 108 |
+
font-size: 17px;
|
| 109 |
+
font-weight: 600;
|
| 110 |
+
}
|
| 111 |
+
|
| 112 |
+
.ae-open h3 {
|
| 113 |
+
color: #243907;
|
| 114 |
+
}
|
| 115 |
+
|
| 116 |
+
.ae-prop h3 {
|
| 117 |
+
color: #b83400;
|
| 118 |
+
}
|
| 119 |
+
|
| 120 |
+
.ae-card p {
|
| 121 |
+
margin: 0;
|
| 122 |
+
font-size: 14px;
|
| 123 |
+
line-height: 1.6;
|
| 124 |
+
}
|
| 125 |
+
|
| 126 |
+
.ae-divider {
|
| 127 |
+
border: 0;
|
| 128 |
+
border-top: 1px solid currentColor;
|
| 129 |
+
opacity: .18;
|
| 130 |
+
margin: 16px 0 12px 0;
|
| 131 |
+
}
|
| 132 |
+
|
| 133 |
+
.ae-links {
|
| 134 |
+
font-size: 13px;
|
| 135 |
+
}
|
| 136 |
+
|
| 137 |
+
.ae-links a {
|
| 138 |
+
color: inherit;
|
| 139 |
+
text-decoration: underline;
|
| 140 |
+
}
|
| 141 |
+
|
| 142 |
+
.ae-releases {
|
| 143 |
+
background: #242424;
|
| 144 |
+
color: #e1e1e1;
|
| 145 |
+
border: 1.5px solid #3d3d3d;
|
| 146 |
+
border-radius: 14px;
|
| 147 |
+
padding: 22px 26px;
|
| 148 |
+
margin-bottom: 22px;
|
| 149 |
+
}
|
| 150 |
+
|
| 151 |
+
.ae-badge-release {
|
| 152 |
+
background: #e1e1e1;
|
| 153 |
+
color: #242424;
|
| 154 |
+
}
|
| 155 |
+
|
| 156 |
+
.ae-release-item {
|
| 157 |
+
display: flex;
|
| 158 |
+
gap: 12px;
|
| 159 |
+
align-items: flex-start;
|
| 160 |
+
}
|
| 161 |
+
|
| 162 |
+
.ae-dot {
|
| 163 |
+
width: 8px;
|
| 164 |
+
height: 8px;
|
| 165 |
+
border-radius: 999px;
|
| 166 |
+
background: #e1e1e1;
|
| 167 |
+
margin-top: 8px;
|
| 168 |
+
flex: 0 0 auto;
|
| 169 |
+
}
|
| 170 |
+
|
| 171 |
+
.ae-release-item h3 {
|
| 172 |
+
margin: 0 0 6px 0;
|
| 173 |
+
font-size: 17px;
|
| 174 |
+
color: #e1e1e1;
|
| 175 |
+
}
|
| 176 |
+
|
| 177 |
+
.ae-release-item p {
|
| 178 |
+
margin: 0 0 12px 0;
|
| 179 |
+
font-size: 14px;
|
| 180 |
+
line-height: 1.6;
|
| 181 |
+
color: #cfcfcf;
|
| 182 |
+
}
|
| 183 |
+
|
| 184 |
+
.ae-pill {
|
| 185 |
+
display: inline-flex;
|
| 186 |
+
align-items: center;
|
| 187 |
+
gap: 6px;
|
| 188 |
+
padding: 5px 14px;
|
| 189 |
+
margin-right: 8px;
|
| 190 |
+
margin-bottom: 8px;
|
| 191 |
+
border-radius: 999px;
|
| 192 |
+
background: #3d3d3d;
|
| 193 |
+
border: 1px solid #555555;
|
| 194 |
+
color: #f2f2f2 !important;
|
| 195 |
+
text-decoration: none !important;
|
| 196 |
+
font-size: 13px;
|
| 197 |
+
}
|
| 198 |
+
|
| 199 |
+
.ae-fr {
|
| 200 |
+
margin-top: 28px;
|
| 201 |
+
padding-top: 22px;
|
| 202 |
+
border-top: 1px solid #3d3d3d;
|
| 203 |
+
}
|
| 204 |
+
|
| 205 |
+
.ae-fr-title {
|
| 206 |
+
margin: 0 0 16px 0;
|
| 207 |
+
font-size: 18px;
|
| 208 |
+
color: #d9e7ff;
|
| 209 |
+
}
|
| 210 |
+
|
| 211 |
+
@media (max-width: 640px) {
|
| 212 |
+
.ae-grid {
|
| 213 |
+
grid-template-columns: 1fr;
|
| 214 |
+
}
|
| 215 |
+
|
| 216 |
+
.ae-root {
|
| 217 |
+
padding: 18px 8px;
|
| 218 |
+
}
|
| 219 |
+
}
|
| 220 |
</style>
|
| 221 |
|
| 222 |
+
<div class="ae-root">
|
| 223 |
+
<div class="ae-lang">
|
| 224 |
+
<span class="active">🇬🇧 EN</span>
|
| 225 |
+
<span>🇫🇷 FR below</span>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 226 |
</div>
|
| 227 |
|
| 228 |
+
<section class="ae-mission">
|
| 229 |
+
<div class="ae-badge ae-badge-blue">🎯 Our mission</div>
|
| 230 |
+
<h2 class="ae-title">Building efficient AI models</h2>
|
| 231 |
+
<p class="ae-text">
|
| 232 |
+
At AlphaEdge, we aim to offer AI models that can run on any type of hardware in a flexible way,
|
| 233 |
+
via API or Edge, on GPU as well as CPU. Our goal is to deliver high performance on complex tasks
|
| 234 |
+
while significantly reducing latency, memory consumption, and inference costs. We expose this
|
| 235 |
+
vision through two channels: our open models and our proprietary models.
|
| 236 |
+
</p>
|
| 237 |
+
</section>
|
| 238 |
|
| 239 |
+
<section class="ae-grid">
|
| 240 |
+
<div class="ae-card ae-open">
|
| 241 |
+
<div class="ae-badge ae-badge-open">🔓 Open source</div>
|
| 242 |
+
<h3>Open Models</h3>
|
| 243 |
+
<p>
|
| 244 |
+
The open source models available on this Hugging Face organization are based on existing ones,
|
| 245 |
+
published under permissive licenses, for which we propose improvements. The goal here is to
|
| 246 |
+
showcase our expertise, particularly in state-of-the-art compression techniques, applied to
|
| 247 |
+
model classes well known to the community.
|
| 248 |
+
</p>
|
| 249 |
+
<hr class="ae-divider">
|
| 250 |
+
<div class="ae-links">
|
| 251 |
+
<a href="https://huggingface.co/alphaedge-ai/models">Available models</a>
|
| 252 |
+
·
|
| 253 |
+
<a href="https://huggingface.co/alphaedge-ai/datasets">Available datasets</a>
|
| 254 |
</div>
|
|
|
|
|
|
|
|
|
|
|
|
|
| 255 |
</div>
|
| 256 |
|
| 257 |
+
<div class="ae-card ae-prop">
|
| 258 |
+
<div class="ae-badge ae-badge-prop">🔒 Proprietary</div>
|
| 259 |
+
<h3>Commercial Models</h3>
|
| 260 |
+
<p>
|
| 261 |
+
Our proprietary models are built on a new architecture that we call ELM
|
| 262 |
+
(Efficient Language Models). They offer optimized performance for professional use cases
|
| 263 |
+
requiring full sovereignty and real-time usage. Resource-efficient, they are available
|
| 264 |
+
through our API or can be deployed on-premises on your hardware at your company.
|
| 265 |
+
</p>
|
| 266 |
+
<hr class="ae-divider">
|
| 267 |
+
<div class="ae-links">
|
| 268 |
+
<a href="https://alphaedge-ai.com/">Our website</a>
|
| 269 |
</div>
|
|
|
|
|
|
|
|
|
|
|
|
|
| 270 |
</div>
|
| 271 |
+
</section>
|
| 272 |
|
| 273 |
+
<section class="ae-releases">
|
| 274 |
+
<div class="ae-badge ae-badge-release">🗃️ Our main open-source releases</div>
|
| 275 |
+
|
| 276 |
+
<div class="ae-release-item">
|
| 277 |
+
<div class="ae-dot"></div>
|
| 278 |
<div>
|
| 279 |
+
<h3>Trimming</h3>
|
| 280 |
+
<p>
|
| 281 |
+
A key technique for effectively reducing the size of language models while preserving their
|
| 282 |
+
performance, making them easier to deploy. We invite you to read the blog post written on
|
| 283 |
+
the subject to learn about all the benefits of this technique, and to explore the Space to
|
| 284 |
+
browse the more than 5,000 models based on this technique that we propose.
|
| 285 |
+
</p>
|
| 286 |
+
<a class="ae-pill" href="https://alphaedge-ai.com/">📖 Blog post</a>
|
| 287 |
+
<a class="ae-pill" href="https://huggingface.co/spaces/alphaedge-ai/README">🤗 HF Space</a>
|
| 288 |
</div>
|
| 289 |
</div>
|
| 290 |
+
|
| 291 |
+
<div class="ae-fr">
|
| 292 |
+
<h2 class="ae-fr-title">Version française</h2>
|
| 293 |
+
|
| 294 |
+
<p>
|
| 295 |
+
Chez AlphaEdge, nous souhaitons proposer des modèles d'IA pouvant s'exécuter sur tout type de
|
| 296 |
+
support de manière flexible, via API ou Edge, sur GPU comme sur CPU. Notre but est d'offrir des
|
| 297 |
+
performances élevées sur des tâches complexes tout en réduisant fortement la latence, la
|
| 298 |
+
consommation mémoire et les coûts d'inférence.
|
| 299 |
+
</p>
|
| 300 |
+
|
| 301 |
+
<p>
|
| 302 |
+
Nos modèles open source s'appuient sur des modèles existants publiés sous licences permissives,
|
| 303 |
+
auxquels nous apportons des améliorations. Nos modèles propriétaires reposent sur une nouvelle
|
| 304 |
+
architecture que nous appelons ELM, pour Efficient Language Models.
|
| 305 |
+
</p>
|
| 306 |
+
</div>
|
| 307 |
+
</section>
|
| 308 |
+
</div>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|