Groq runs open-weight models on custom LPU silicon and posts the lowest latency numbers in the industry. For agents where the round-trip matters more than the marginal quality (voice, real-time triage, latency-bound UX), Groq is hard to beat. The model catalogue is curated rather than comprehensive.
app.yaml1brain:2 provider: openai_compat3 model: llama-3.3-70b-versatile4 config:5 base_url: "https://api.groq.com/openai/v1"6 api_key: "{{env.GROQ_API_KEY}}"Groqdoesn't have a first-class catalog entry yet. Configure it inline using env templates, the same way the blog examples show. Native catalog support is on the roadmap.
GROQ_API_KEYGroq Cloud API key# add to ~/.digitorn/.env
GROQ_API_KEY=...# 1. install the runtime
curl -sSL https://digitorn.ai/install | sh
# 2. drop your Groq key in the env file
echo 'GROQ_API_KEY=...' >> ~/.digitorn/.env
# 3. install a starter agent and chat
digitorn install hub://digitorn/digitorn-code
digitorn chat digitorn-codeEngineering notes from the Digitorn team. No marketing, no launch announcements, no "10 prompts that will change your life". Just the things we write that we'd want to read.