# leptonai/leptonai

A Pythonic framework to simplify AI service building

Repository: https://github.com/leptonai/leptonai
Canonical: https://ross.abutalabs.com/products/leptonai
Homepage: https://lepton.ai/
Language: Python
License: Apache-2.0
License Family: permissive
Topics: artificial-intelligence, cloud, deep-learning, gpu, machine-learning, python
Last push: 2026-08-26T00:35:59+00:00

## Health v2 (maintenance only)
Score: 94/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 96, longevity 78
- inputs: {"age_days": 1092, "days_push": 8, "days_rel": 26, "gap_med": 10.5, "n_releases_24m": 41}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2825, forks 196 (observed 2026-08-28T04:07:24.138693+00:00)

## What it is
LeptonAI is a Python library and `lep` CLI for operating NVIDIA DGX Cloud Lepton, a platform that unifies global GPU compute for AI development. It lets developers deploy endpoints, run batch jobs, launch dev pods, and call deployed services like native Python functions.

## Use cases
- deploy a model endpoint to cloud GPUs from Python
- run batch training jobs on remote GPU clusters
- spin up interactive GPU dev pods for experimentation
- call a deployed AI endpoint like a local Python function
- manage secrets, storage, and fine-tuning jobs from the CLI
- let AI agents operate GPU cloud infrastructure via skills

## When to choose
- you use NVIDIA DGX Cloud Lepton and want a Pythonic interface
- you need to deploy and manage AI services across multi-cloud GPU compute
- you want CLI and SDK access to endpoints, jobs, and pods in one tool

## When to avoid
- you need a self-hosted, vendor-neutral deployment framework
- you don't use NVIDIA DGX Cloud Lepton or NVIDIA GPU cloud services
- you only need local model inference without cloud orchestration

## Facets
- artifact type: library
- maturity: active
- function: sdk, cli, deployment, cloud, llm-inference
- domain: artificial-intelligence, machine-learning, cloud-computing, gpu-computing, developer-tools
- platform: python, cli, cloud
- tags: nvidia-dgx-cloud, gpu-compute, ai-service-deployment, lep-cli, agent-skills

## Member repositories
- leptonai/leptonai (main) score 94

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:24.138693+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T07:38:14.744247+00:00, confidence not recorded.
  - readme: https://github.com/leptonai/leptonai (fetched 2026-08-28T04:07:24.138693+00:00, sha 110fc21f8087)
  - homepage: https://lepton.ai/ (fetched 2026-08-29T09:53:47.510566+00:00, sha 9be85a1bcc76)
  - site_page: https://www.nvidia.com/en-us/about-nvidia/terms-of-service (fetched 2026-08-29T09:53:47.524082+00:00, sha 85469b6ff0a1)
  - site_page: https://www.nvidia.com/en-us/about-nvidia/accessibility (fetched 2026-08-29T09:53:47.526448+00:00, sha 8f61ce4143ec)
  - site_page: https://www.nvidia.com/en-us/about-nvidia (fetched 2026-08-29T09:53:47.515817+00:00, sha 33e0c25ebeb0)
  - site_page: https://www.nvidia.com/en-us/about-nvidia/careers (fetched 2026-08-29T09:53:47.517599+00:00, sha 1c3b99bfdf98)
  - site_page: https://www.nvidia.com/en-us/about-nvidia/privacy-policy (fetched 2026-08-29T09:53:47.519460+00:00, sha 5362c58d0750)
  - site_page: https://www.nvidia.com/en-us/about-nvidia/privacy-center (fetched 2026-08-29T09:53:47.521843+00:00, sha b098377da9cc)
  - site_page: https://www.nvidia.com/en-us/about-nvidia/company-policies (fetched 2026-08-29T09:53:47.528011+00:00, sha b169611fd7d6)
  - site_page: https://resources.nvidia.com/l/en-us-gpu (fetched 2026-08-29T09:53:47.513416+00:00, sha 3dee1a40418d)
- Data as of 2026-08-30T08:39:29.467469+00:00.