Skip to content

Commit e8dcfe1

Browse files
peterschmidt85peterschmidt85
committed
Deploying to gh-pages from @ dstackai/dstack@42c0752 🚀
1 parent a33f3c8 commit e8dcfe1

File tree

5 files changed

+218
-215
lines changed

5 files changed

+218
-215
lines changed

‎assets/stylesheets/landing.css‎

Lines changed: 2 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -66,7 +66,8 @@
6666

6767
@media screen and (max-width: 76.1875em) {
6868
.tx-landing h1 {
69-
font-size: 1.4rem;
69+
font-size: 1.9rem;
70+
margin: 24px -24px;
7071
}
7172

7273
.tx-landing__hero_text {

‎index.html‎

Lines changed: 43 additions & 42 deletions
Original file line numberDiff line numberDiff line change
@@ -3952,9 +3952,9 @@
39523952
<h1>The new GPU-native orchestration</h1>
39533953

39543954
<p>
3955-
<span class="highlight">dstack</span> provides ML teams with a unified control plane for GPU provisioning and orchestration
3956-
across cloud, Kubernetes, and on-prem. It streamlines development, training, and inference — reducing costs 3–7x and
3957-
preventing lock-in.
3955+
<span class="highlight">dstack</span> is a GPU-native orchestration built for modern AI teams. It simplifies
3956+
GPU provisioning and workload management across clouds, Kubernetes, and on-prem —
3957+
through a single unified control plane.
39583958
</p>
39593959
</div>
39603960

@@ -3981,22 +3981,21 @@ <h1>The new GPU-native orchestration</h1>
39813981
</div>
39823982
</div>
39833983
<div class="tx-landing__hero_button_container_text" style="text-align: center; font-size: 85%; margin-top: 1.25rem">
3984-
One control plane for GPUs across cloud, Kubernetes, and on-prem.
3984+
Reduce GPU costs by 3–7× and eliminate vendor lock-in.
39853985
</div>
39863986
</div>
39873987

39883988
<div class="tx-landing__major_feature">
39893989
<div class="section">
39903990
<div class="block margin right">
3991-
<h2>An open platform for GPU orchestration</h2>
3991+
<h2>A unified control plane for GPU orchestration</h2>
39923992
<p>
3993-
Managing AI infrastructure requires efficient GPU orchestration, whether workloads run
3994-
on a single GPU cloud, across multiple GPU providers, or on-prem clusters.
3993+
Managing AI infrastructure requires efficient GPU orchestration tightly integrated with open-source training and
3994+
inference frameworks.
39953995
</p>
39963996

39973997
<p>
3998-
<span class="highlight">dstack</span> provides an open stack for GPU orchestration that streamlines development, training,
3999-
and inference, and can be used with any hardware, open-source tools, and frameworks.
3998+
<span class="highlight">dstack</span> provides a unified control plane&mdash;so workloads stay portable, reproducible, and infrastructure remains interchangeable.
40003999
</p>
40014000

40024001
<!-- TODO: Add `Why dstack?` -->
@@ -4072,22 +4071,21 @@ <h2>An open platform for GPU orchestration</h2>
40724071
<div class="block margin">
40734072
<h2>Native integration with GPU clouds</h2>
40744073
<p>
4075-
<span class="highlight">dstack</span> natively integrates with top GPU clouds—automating cluster provisioning and
4076-
workload orchestration to maximize efficiency and minimize overhead.
4074+
<span class="highlight">dstack</span> natively integrates with leading GPU clouds for fast, efficient provisioning.
40774075
</p>
40784076

40794077
<p>
4080-
It can provision and manage VM clusters through native integrations or via <span class="highlight">Kubernetes</span>.
4078+
It can provision and manage GPU VMs directly through cloud APIs, or operate through <span class="highlight">Kubernetes</span> when required.
40814079
</p>
40824080

40834081
<p>
40844082
<a href="/docs/concepts/backends" target="_blank" class="md-button md-button-secondary">
40854083
<span>Backends</span>
4086-
<span class="icon"><svg viewBox="0 0 13 10" xmlns="http://www.w3.org/2000/svg">
4084+
<!-- <span class="icon"><svg viewBox="0 0 13 10" xmlns="http://www.w3.org/2000/svg">
40874085
<path
40884086
d="M12.823 4.164L8.954.182a.592.592 0 0 0-.854 0 .635.635 0 0 0 0 .88l2.836 2.92H.604A.614.614 0 0 0 0 4.604c0 .344.27.622.604.622h10.332L8.1 8.146a.635.635 0 0 0 0 .88.594.594 0 0 0 .854 0l3.869-3.982a.635.635 0 0 0 0-.88z"
40894087
fill-rule="nonzero" fill="currentColor" class="fill-main"></path>
4090-
</svg></span>
4088+
</svg></span> -->
40914089
</a>
40924090
</p>
40934091
</div>
@@ -4100,35 +4098,32 @@ <h2>Native integration with GPU clouds</h2>
41004098
<h2>Easy to use with on-prem clusters</h2>
41014099

41024100
<p>
4103-
For provisioned <span class="highlight">Kubernetes</span> clusters, connect them to <span class="highlight">dstack</span> using the <span
4104-
class="highlight">Kubernetes</span> backend.
4105-
If you run vanilla bare-metal servers or VMs without <span class="highlight">Kubernetes</span>, use <span class="highlight">SSH fleets</span>
4106-
instead.
4101+
If you already run <span class="highlight">Kubernetes</span> on-prem, connect your cluster to <span class="highlight">dstack</span> using the <span class="highlight">Kubernetes</span> backend.
4102+
</p>
4103+
4104+
<p>
4105+
For bare-metal servers or VMs without <span class="highlight">Kubernetes</span>, use <span class="highlight">SSH fleets</span> to orchestrate GPUs directly.
41074106
</p>
41084107

41094108
<p>
41104109
<a href="/docs/concepts/backends#ssh-fleets" target="_blank" class="md-button md-button-secondary">
41114110
<span>SSH fleets</span>
4112-
<span class="icon"><svg viewBox="0 0 13 10" xmlns="http://www.w3.org/2000/svg">
4111+
<!-- <span class="icon"><svg viewBox="0 0 13 10" xmlns="http://www.w3.org/2000/svg">
41134112
<path
41144113
d="M12.823 4.164L8.954.182a.592.592 0 0 0-.854 0 .635.635 0 0 0 0 .88l2.836 2.92H.604A.614.614 0 0 0 0 4.604c0 .344.27.622.604.622h10.332L8.1 8.146a.635.635 0 0 0 0 .88.594.594 0 0 0 .854 0l3.869-3.982a.635.635 0 0 0 0-.88z"
41154114
fill-rule="nonzero" fill="currentColor" class="fill-main"></path>
4116-
</svg></span>
4115+
</svg></span> -->
41174116
</a>
41184117

41194118
<a href="/docs/concepts/backends#kubernetes" target="_blank" class="md-button md-button-secondary">
41204119
<span>Kubernetes</span>
4121-
<span class="icon"><svg viewBox="0 0 13 10" xmlns="http://www.w3.org/2000/svg">
4120+
<!-- <span class="icon"><svg viewBox="0 0 13 10" xmlns="http://www.w3.org/2000/svg">
41224121
<path
41234122
d="M12.823 4.164L8.954.182a.592.592 0 0 0-.854 0 .635.635 0 0 0 0 .88l2.836 2.92H.604A.614.614 0 0 0 0 4.604c0 .344.27.622.604.622h10.332L8.1 8.146a.635.635 0 0 0 0 .88.594.594 0 0 0 .854 0l3.869-3.982a.635.635 0 0 0 0-.88z"
41244123
fill-rule="nonzero" fill="currentColor" class="fill-main"></path>
4125-
</svg></span>
4124+
</svg></span> -->
41264125
</a>
41274126
</p>
4128-
4129-
<p>
4130-
Either way, connecting existing on-prem clusters to <span class="highlight">dstack</span> takes just minutes.
4131-
</p>
41324127
</div>
41334128

41344129
<div class="block large">
@@ -4149,18 +4144,19 @@ <h2>Easy to use with on-prem clusters</h2>
41494144
<div class="block margin left">
41504145
<h2>Dev environments</h2>
41514146
<p>
4152-
Before training or deployment, ML engineers explore and debug their code.
4147+
Before training or deploying models, ML engineers need interactive GPU access to experiment and debug.
41534148
</p>
41544149

4155-
<p><span class="highlight">dstack</span>'s dev environments make it easy to connect your
4156-
desktop IDE to powerful cloud or on-prem GPUs—streamlining the entire development loop.
4150+
<p><span class="highlight">dstack</span>'s dev environments let you connect desktop IDEs such as <span class="highlight">VS
4151+
Code</span>, <span class="highlight">Cursor</span>, and
4152+
<span class="highlight">Windsurf</span> directly to cloud or on-prem GPUs.
41574153
</p>
41584154

41594155
<p>
41604156
<a href="/docs/dev-environments" target="_blank"
41614157
class="md-button md-button-secondary">
41624158
<span>Dev environments</span>
4163-
<span class="icon"><svg viewBox="0 0 13 10" xmlns="http://www.w3.org/2000/svg"><path d="M12.823 4.164L8.954.182a.592.592 0 0 0-.854 0 .635.635 0 0 0 0 .88l2.836 2.92H.604A.614.614 0 0 0 0 4.604c0 .344.27.622.604.622h10.332L8.1 8.146a.635.635 0 0 0 0 .88.594.594 0 0 0 .854 0l3.869-3.982a.635.635 0 0 0 0-.88z" fill-rule="nonzero" fill="currentColor" class="fill-main"></path></svg></span>
4159+
<!-- <span class="icon"><svg viewBox="0 0 13 10" xmlns="http://www.w3.org/2000/svg"><path d="M12.823 4.164L8.954.182a.592.592 0 0 0-.854 0 .635.635 0 0 0 0 .88l2.836 2.92H.604A.614.614 0 0 0 0 4.604c0 .344.27.622.604.622h10.332L8.1 8.146a.635.635 0 0 0 0 .88.594.594 0 0 0 .854 0l3.869-3.982a.635.635 0 0 0 0-.88z" fill-rule="nonzero" fill="currentColor" class="fill-main"></path></svg></span> -->
41644160
</a>
41654161
</p>
41664162
</div>
@@ -4173,18 +4169,19 @@ <h2>Dev environments</h2>
41734169
<h2>Single-node &amp; distributed tasks</h2>
41744170

41754171
<p>
4176-
Move from single-instance experiments to multi-node distributed training without friction. <span
4177-
class="highlight">dstack</span> lets you define complex jobs with a simple configuration,
4178-
handling the scheduling and orchestration for you.
4172+
Run training or batch workloads on a single GPU, or scale to multi-GPU and multi-node clusters using simple task configurations.
4173+
<span class="highlight">dstack</span> automates cluster provisioning, resource allocation, and job scheduling.
41794174
</p>
41804175

4181-
<p>This allows your team to focus on research while ensuring that expensive cluster resources are utilized efficiently.</p>
4176+
<p>
4177+
During execution, <span class="highlight">dstack</span> reports GPU utilization, memory usage, and GPU health metrics for each job.
4178+
</p>
41824179

41834180
<p>
41844181
<a href="/docs/concepts/tasks" target="_blank"
41854182
class="md-button md-button-secondary">
41864183
<span>Tasks</span>
4187-
<span class="icon"><svg viewBox="0 0 13 10" xmlns="http://www.w3.org/2000/svg"><path d="M12.823 4.164L8.954.182a.592.592 0 0 0-.854 0 .635.635 0 0 0 0 .88l2.836 2.92H.604A.614.614 0 0 0 0 4.604c0 .344.27.622.604.622h10.332L8.1 8.146a.635.635 0 0 0 0 .88.594.594 0 0 0 .854 0l3.869-3.982a.635.635 0 0 0 0-.88z" fill-rule="nonzero" fill="currentColor" class="fill-main"></path></svg></span>
4184+
<!-- <span class="icon"><svg viewBox="0 0 13 10" xmlns="http://www.w3.org/2000/svg"><path d="M12.823 4.164L8.954.182a.592.592 0 0 0-.854 0 .635.635 0 0 0 0 .88l2.836 2.92H.604A.614.614 0 0 0 0 4.604c0 .344.27.622.604.622h10.332L8.1 8.146a.635.635 0 0 0 0 .88.594.594 0 0 0 .854 0l3.869-3.982a.635.635 0 0 0 0-.88z" fill-rule="nonzero" fill="currentColor" class="fill-main"></path></svg></span> -->
41884185
</a>
41894186
</p>
41904187
</div>
@@ -4207,31 +4204,35 @@ <h2>Single-node &amp; distributed tasks</h2>
42074204
<h2>Scalable model inference</h2>
42084205

42094206
<p>
4210-
With <span class="highlight">dstack</span>, you can easily deploy any model as a secure,
4211-
auto-scaling OpenAI-compatible endpoint, all while using your custom code, Docker image, and
4212-
serving framework.
4207+
With <span class="highlight">dstack</span>, you can deploy models as secure,
4208+
auto-scaling, OpenAI-compatible endpoints, integrating with top open-source serving frameworks.
4209+
</p>
4210+
4211+
<p>
4212+
<span class="highlight">dstack</span> supports disaggregated prefill/decode and cache-aware routing, providing
4213+
production-grade, optimized inference.
42134214
</p>
42144215

42154216
<p>
42164217
<a href="/docs/concepts/services" target="_blank"
42174218
class="md-button md-button-secondary">
42184219
<span>Services</span>
4219-
<span class="icon"><svg viewBox="0 0 13 10" xmlns="http://www.w3.org/2000/svg"><path d="M12.823 4.164L8.954.182a.592.592 0 0 0-.854 0 .635.635 0 0 0 0 .88l2.836 2.92H.604A.614.614 0 0 0 0 4.604c0 .344.27.622.604.622h10.332L8.1 8.146a.635.635 0 0 0 0 .88.594.594 0 0 0 .854 0l3.869-3.982a.635.635 0 0 0 0-.88z" fill-rule="nonzero" fill="currentColor" class="fill-main"></path></svg></span>
4220+
<!-- <span class="icon"><svg viewBox="0 0 13 10" xmlns="http://www.w3.org/2000/svg"><path d="M12.823 4.164L8.954.182a.592.592 0 0 0-.854 0 .635.635 0 0 0 0 .88l2.836 2.92H.604A.614.614 0 0 0 0 4.604c0 .344.27.622.604.622h10.332L8.1 8.146a.635.635 0 0 0 0 .88.594.594 0 0 0 .854 0l3.869-3.982a.635.635 0 0 0 0-.88z" fill-rule="nonzero" fill="currentColor" class="fill-main"></path></svg></span> -->
42204221
</a>
42214222

4222-
<a href="/docs/concepts/gateways" target="_blank"
4223+
<!-- <a href="/docs/concepts/gateways" target="_blank"
42234224
class="md-button md-button-secondary">
42244225
<span>Gateways</span>
42254226
<span class="icon"><svg viewBox="0 0 13 10" xmlns="http://www.w3.org/2000/svg"><path d="M12.823 4.164L8.954.182a.592.592 0 0 0-.854 0 .635.635 0 0 0 0 .88l2.836 2.92H.604A.614.614 0 0 0 0 4.604c0 .344.27.622.604.622h10.332L8.1 8.146a.635.635 0 0 0 0 .88.594.594 0 0 0 .854 0l3.869-3.982a.635.635 0 0 0 0-.88z" fill-rule="nonzero" fill="currentColor" class="fill-main"></path></svg></span>
4226-
</a>
4227+
</a> -->
42274228
</p>
42284229
</div>
42294230
</div>
42304231
</div>
42314232

42324233
<div class="tx-landing__trusted_by">
42334234
<div class="tx-landing__highlights_text">
4234-
<h2>Loved by world-class ML teams</h2>
4235+
<h2>Loved by world-class AI teams</h2>
42354236

42364237
<div class="tx-landing__quotes_grid">
42374238
<div class="cell">

0 commit comments

Comments
 (0)