Bring your own key
Very fast answers:
your Groq key in wuup
Add a Groq API key and wuup answers with open models such as Llama and GPT-OSS, running on Groq's fast inference hardware, so answers stream back quickly. Groq bills you directly; wuup adds no fee and no question limit.
- 1
Create a Groq API key
Sign in to the GroqCloud console and create a new secret key. Give it a name like "wuup" so you can revoke it later.
- 2
Open Settings, then API key
In wuup, open Settings and choose the API key section.
- 3
Choose Groq and paste the key
wuup checks the key with the provider before saving it, and shows it masked from then on.
- 4
Pick a model and ask
Choose a model from the list for your provider. Questions are no longer counted against the free tier.
What to know
What changes when you use your own key
Your questions are answered by Groq models you choose, billed by Groq to your account at their API prices. wuup adds no charge and no markup, and the 50-question free tier limit no longer applies.
Everything else stays the same: the same read-only guard on MongoDB, the same citations for documents, the same charts built from real rows and the same queries-used panel.
What is sent to the provider
To answer a question, wuup sends the provider your question, the recent chat context, a schema summary of the sources in the chat (with sensitive fields redacted), and the rows or passages a query returned. Your connection strings and the key itself are never sent in a prompt.
Check Groq's API data policy for how they handle API traffic. You can remove your key in Settings at any time to return to the free tier.
Which model Auto picks on Groq
When you save the key, wuup asks Groq for its model list and keeps the chat models. Speech, text to speech and safety models are left out, because they can't answer questions about your data.
Auto uses the strongest tool-calling model your key can see, and a small, fast model for chat titles. If Groq rate limits a model, Auto moves on to the next one, and the model menu shows when the limited model is free again.
Related
Other providers
Questions,
answered
The chat models your key can see. wuup reads them from Groq when you save the key and lists them in the model menu, with Auto picking the best one for tool calling.
Groq runs models on its own inference hardware built for fast token generation, so answers start and finish quickly. Speed varies with the model and with Groq's load.
Groq bills by tokens at the rates on groq.com/pricing, and its free plan has rate limits. wuup adds nothing on top.