Back to search
Top Contributors
1.2k
commits
905
commits
333
commits
234
commits
208
commits
181
commits
127
commits
47
commits
15
commits
7
commits
Control plane for agents and engineers to provision compute and run training and inference across NVIDIA, AMD, TPU, and Tenstorrent GPUs—on clouds, Kubernetes, and bare-metal clusters.
Stars
2.1k
Forks
222
Watchers
2.1k
Open Issues
85
Overall repository health assessment
No package.json found
This might not be a Node.js project
1.2k
commits
905
commits
333
commits
234
commits
208
commits
181
commits
127
commits
47
commits
15
commits
7
commits
Allow combining on/off idle_duration between runs and fleets (#3756)
4062911View on GitHubAdd case study on Graphsignal's use of dstack for inference benchmarking (#3751)
5b1233cView on GitHubVerda: make startup script and SSH key lifecycle per-instance with reliable cleanup (#3718)
fc9afa9View on GitHubHandle concurrent repo blob/file archive uploads (#3737)
a325f56View on GitHubUpdate Server Deployment guide for pipelines (#3745)
affb2cdView on GitHub