n8n-nodes-nvidia-nim
v— GitHub

NVIDIA NIM

Free AI models that plug straight into n8n.

Community node @sateeshreddy/n8n-nodes-nvidia-nim — chat and embeddings as native AI sub-nodes, no HTTP gymnastics.

Install in n8n View on npm

The issue

n8n locks you into paid embedding nodes.

Every vector store in n8n only accepts a native Embeddings sub-node. An HTTP Request cannot fill that slot. Chat models for Agents have the same constraint. Paid APIs add up fast while you are still prototyping.

  • No free native embeddings Pinecone, Qdrant, Supabase, Weaviate, and PGVector all require an official embedding sub-node — not a workaround.
  • Manual NIM wiring is brittle HTTP nodes, auth headers, and model IDs work until you need Agents, memory, or RAG — then the seams show.
  • Cost before product-market fit OpenAI, Anthropic, and Gemini burn budget during experiments. NVIDIA NIM already hosts strong models on a free developer tier.

What this package fixes

Two native nodes. One credential. Free NIM models.

Drop into AI Agent, Basic LLM Chain, or any vector store — the same way OpenAI or Anthropic nodes work, without the invoice.

NVIDIA NIM

Language model sub-node for Agents and LLM chains. Smart defaults set temperature and max tokens per model.

17+ chat models

NVIDIA NIM Embeddings

Native embeddings sub-node for Qdrant, Pinecone, PGVector, Supabase, Weaviate, and Chroma.

5 embedding models

Install

Live in under a minute.

  1. 01

    Community Nodes

    In n8n: Settings → Community Nodes → Install

    @sateeshreddy/n8n-nodes-nvidia-nim
  2. 02

    Add your API key

    Create a free key at build.nvidia.com, then Credentials → NVIDIA NIM API.

  3. 03

    Wire the sub-node

    Search “NVIDIA NIM” in a Model or Embeddings slot and run.

Live from npm

Version & what changed

Package version and publish history load automatically from the npm registry. Release notes prefer GitHub Releases or CHANGELOG when present.

What this release addresses

Loading release notes…