llmpm — NPM for LLMs | Install, run & ship AI Models

1 min read Original article ↗

Now available on PyPI & npm

llmpm — LLM Package Manager

LLM PACKAGE MANAGER

NPM for LLMs  |  Install, run & ship AI Models

Download and run 10,000+ free models from HuggingFace, Ollama & Mistral with a single command.

$pip install llmpm

$npm install -g llmpm

Browse ModelsView Docs

QUICK START

From install to running a model in under a minute.

#Install a model

$llmpm install meta-llama/Llama-3.2-3B-Instruct

$

#Run it interactively

$llmpm run meta-llama/Llama-3.2-3B-Instruct

$

#Or serve as OpenAI-compatible API

$llmpm serve meta-llama/Llama-3.2-3B-Instruct

SEE IT IN ACTION

Watch llmpm install and run a model end-to-end in seconds.

COMMANDS

Everything you need to manage large language models.

Install Models

Install any model from HuggingFace, Ollama, or Mistral with a single command.

$llmpm install meta-llama/Llama-3.2-3B-Instruct

Run Instantly

Auto-detects model type and launches the right backend. Text, image, audio — it just works.

$llmpm run meta-llama/Llama-3.2-3B-Instruct

Serve as API

Expose any model as an OpenAI-compatible REST API with a built-in chat UI.

$llmpm serve meta-llama/Llama-3.2-3B-Instruct

Push Models

Upload fine-tuned models directly to HuggingFace Hub with one command.

$llmpm push my-org/my-fine-tune

SUPPORTED MODEL TYPES

One tool for every modality.

💬

Text Generation

GGUF via llama.cpp & Transformer checkpoints

🎨

Text-to-Image

Diffusion models via Diffusers

👁️

Vision Models

Image-to-text, captioning & VQA

🎤

Speech-to-Text

ASR with Whisper and more

🔊

Text-to-Speech

TTS generation to WAV output

Start in 30 seconds.

No config files, no complex setup. Just install and run.