Provider · Groq

Linda + Groq.

Groq's LPUs serve open-weight models at ~600+ tokens/second. Best when you want every keystroke to feel instant.

Recommended models

llama-3.3-70b-versatile — the workhorse open model.
qwen-2.5-32b — strong tool calling.

Config

new Linda({
  transport: {
    mode: "browser",
    provider: "groq",
    apiKey: import.meta.env.PUBLIC_GROQ_KEY,
    model: "llama-3.3-70b-versatile",
  },
});

When to pick Groq

You need sub-second time-to-first-token.
Your flow is short — long contexts not your bottleneck.
You're OK with open-weight models for the task.

Get an API key at console.groq.com.

Ship an agent-driven flow this afternoon.

Install Linda, paste a config, and your form turns into an agent that fills its own inputs.

Install Linda Browse use cases