Groq

Cloud API provider for fast inference on Groq LPU hardware via api.groq.com.

1 min read Provider Backend cloud openai-api low-latency

Groq provides extremely fast inference using custom LPU hardware. The API is OpenAI-compatible at api.groq.com.

[backend]
name = "groq"
url = "https://api.groq.com"
model = "llama-3.3-70b-versatile"
api_key = "gsk_..."

Groq uses the /openai/v1/chat/completions path prefix (not /v1/chat/completions).

POST https://api.groq.com/openai/v1/chat/completions

Related Documentation