Free
Free
Open-source npm package; documentation and CLI examples use npx without listing a paid subscription.
- Access
- Open-source npm package; documentation and CLI examples use npx without listing a paid subscription.
node-llama-cpp
● Identity checkedCode · Vendor site 2026-10-04
node-llama-cpp is a Node.js package for running large language models locally via llama.cpp bindings. It supports Metal, CUDA, and Vulkan, ships prebuilt binaries, and adds structured JSON output, function calling, embeddings, and CLI tools for chat and hardware inspection.
- Category
- Code
- Free access
- Open-source npm package; documentation and CLI examples use npx without listing a paid subscription.
- API access
- Published API reference and CLI cover programmatic model use, JSON schema enforcement, and function calling from Node.js projects.
Is node-llama-cpp right for you?
A good fit for
Developers building local or self-hosted LLM features in Node.js, Bun, or Electron who want llama.cpp ergonomics.
Before you choose
Performance and supported models depend on local GPU or CPU hardware and the weights you download.
Local LLMs in Node.js
node-llama-cpp wraps llama.cpp for JavaScript runtimes, bundling chat CLI commands, GPU inspection utilities, embeddings, grammar controls, and native binaries so teams can embed local models without maintaining custom C++ bindings.
What it can do
Features & capabilities
Unknown is different from unavailable. Each fact carries its own evidence.
| Capability | Value | Evidence | Checked |
|---|---|---|---|
| Runtime targets | Docs state zero-config defaults and support for Node.js, Bun, and Electron. | Facts sourced | 2026-10-04 |
| Hardware backends | Site highlights Metal, CUDA, and Vulkan support with automatic hardware adaptation. | Facts sourced | 2026-10-04 |
| Structured output | Marketing lists JSON schema constrained generation and on-demand function calling. | Facts sourced | 2026-10-04 |
Understand the total cost
node-llama-cpp pricing & plans
Alternatives to node-llama-cpp
View all ↗
OneCompilerOneCompiler is an online IDE and compiler platform for writing, running, and sharing code in 110+ languages and databases. It adds AI-assisted building blocks, studio time, and team-oriented paid tiers on top of a free compiler that supports unlimited code executions with daily AI credit limits.
Explore tool
Code Snippets AIIt supports local and cloud LLMs, codebase indexing on paid tiers, and a free Basic plan with your own API keys.
Explore toolBolt.new
Bolt.new is a browser-based AI app builder where users describe a web or mobile app in chat and get a running project with hosting, databases and custom domains on paid tiers.
Explore toolThe practical questions
Frequently asked questions
What is node-llama-cpp used for?
node-llama-cpp is a Node.js package for running large language models locally via llama.cpp bindings. It supports Metal, CUDA, and Vulkan, ships prebuilt binaries, and adds structured JSON output, function calling, embeddings, and CLI tools for chat and hardware inspection.
Does node-llama-cpp have a free plan?
Open-source npm package; documentation and CLI examples use npx without listing a paid subscription.. This record lists ongoing free access; check the plan limits before starting.
How much does node-llama-cpp cost?
No paid monthly price is listed; this record treats the product as free to start. See the plan cards for entitlements, billing commitments and seat minimums.
Can I use node-llama-cpp through an API?
Published API reference and CLI cover programmatic model use, JSON schema enforcement, and function calling from Node.js projects.. API access and subscription access may have different terms; consult the linked sources.
What should I check before choosing it?
Performance and supported models depend on local GPU or CPU hardware and the weights you download.
Price history
No retained pricing changes yet. A current price alone does not establish a historical trend.
How this profile is supported
Identity and profile
Reviewed vendor source
Read original source ↗Facts apply to the named version and check date. Send a sourced correction if something changed.