Do I need a Groq API key to use this skill?

Yes, a valid GROQ_API_KEY is required to access Groq's high-speed chat completion models.

Does this skill require an internet connection for embeddings?

No, embeddings are processed locally via Ollama, though you must pull the nomic-embed-text model once during setup.

Which Groq models are supported?

It supports llama-3.3-70b-versatile (default), llama-3.1-8b-instant, gemma2-9b-it, and legacy Llama-3 models.

How does this handle large-scale text embedding?

The skill supports piping input from files or other scripts into the embed.sh utility, making it efficient for batch processing.

Groq & Ollama AI Bridge

Name: Groq & Ollama AI Bridge
Author: lev-os

bylev-os

0•

Data Science & ML

Enables high-speed chat completions via Groq Cloud and local text embeddings through Ollama for efficient RAG workflows.

The groq skill integrates ultra-fast Llama-3 inference with local embedding capabilities to provide a high-performance AI toolkit for Claude. By leveraging Groq's specialized hardware for chat completions and Ollama's local environment for vector embeddings, this skill allows developers to build low-latency applications and Retrieval-Augmented Generation (RAG) systems without relying on external embedding APIs. It is ideal for users who need the speed of Groq's cloud-based LLMs while maintaining a local footprint for data vectorization.