6 engineers on Hacker News describe using Groq first-hand: 3 endorse it, 0 regret it, 1 are mixed.
“We use Groq with gpt-oss-20b, which gives great results and only adds ~250ms to the processing pipeline.”
endorse · production · sgk284 · 2025-12-25 · source ↗ · wrong label?“since I'm using groq for inference, I still haven't even needed to pay a thing, since it fits within the free limits”
endorse · personal · neoromantique · 2025-12-15 · source ↗ · wrong label?“Sometimes groq errors out so there's built-in support for Replicate and Deepgram as well.”
mixed · production · rob · 2025-06-25 · source ↗ · wrong label?“We replaced most our internal integrations with this and everything is great so far.”
endorse · production · anonzzzies · 2024-04-23 · source ↗ · wrong label?“our Llama 8b (via Groq) implementat”
neutral · production · minotauris · 2026-02-15 · source ↗ · wrong label?“I have been using 4.6 on Cerebras (or Groq with other models) since it dropped”
neutral · personal · anonzzzies · 2025-12-22 · source ↗ · wrong label?