#5757th of 80 in Agent platforms
leptonai
Python SDK and lep CLI for managing NVIDIA DGX Cloud Lepton workloads
Documentation ↗
(opens in a new tab)Website ↗
(opens in a new tab)Repository on GitHub ↗
(opens in a new tab)
- Stars
- 2.8k
- License
- Apache-2.0
- Last commit
- Nov 2026
- Last release
- Oct 2026
- Language
- Python
2,790 on 21 Apr 20262,826 on 10 Nov 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
Client library and `lep` command-line tool for the NVIDIA DGX Cloud Lepton platform. It creates and manages endpoints, Dynamo graph deployments (vLLM, SGLang, TensorRT-LLM), batch jobs, dev pods, Ray clusters, and Slurm inspection. A Python `Client` calls deployed endpoints as functions, and a bundled skill lets Claude Code or Codex drive the CLI.
Who it is for: Users of NVIDIA DGX Cloud Lepton who deploy and manage GPU workloads
Strengths
- Single CLI covers endpoints, jobs, dev pods, Ray clusters, Dynamo deployments and Slurm inspection
- Dynamo commands support multi-node and disaggregated prefill/decode with vLLM, SGLang or TensorRT-LLM
- Shell access through workspace API over HTTPS, no SSH keys needed; Teleport SSH also supported
- Ships a Claude Code and Codex plugin that asks confirmation before modifying workloads
Weaknesses
- Only works against NVIDIA DGX Cloud Lepton; it is not a self-hostable serving stack
- Requires a Lepton workspace account and login to do anything useful
- Teleport SSH needs a local tsh v18+ and workspace-side Teleport enablement
- Slurm job submission and cancellation stay outside the CLI; it only inspects
What it needs
- Needs NVIDIA DGX Cloud Lepton workspace, tsh v18+ (optional, for Teleport SSH)
- Models: vLLM, SGLang, TensorRT-LLM
Quick answers
Is leptonai open source?
Yes. Its code is on GitHub under the Apache-2.0 license.
Also in Agent platforms
See all 80| Rank | Project | Score |
|---|---|---|
| 1 | AutoGPTRole-based AI agents coordinated by one lead agent, plus a visual workflow builder | 95 out of 100 |
| 2 | daily_stock_analysisLLM-driven daily analysis of watchlist stocks across six markets, pushed to chat apps | 95 out of 100 |
| 3 | LangflowVisual builder for AI agents and workflows, exposed as API or MCP server | 94 out of 100 |
| #4 | deer-flowAgent harness running sub-agents, memory and sandboxes with pluggable skills | 93 out of 100 |
| #5 | DifyVisual builder for LLM workflows, RAG pipelines and agents, with APIs | 92 out of 100 |