Groq

Groq runs open-source models on custom LPU hardware, delivering fast inference for both transcription and summarization.

Roles: Transcription and Summarization

Role	Model
Transcription	whisper-large-v3
Summarization	llama-3.3-70b-versatile

Note

Groq’s transcription endpoint doesn’t return speaker labels. Quendian can add them on-device if you opt in via Preferences → Speakers → “Identify speakers on-device”. Local diarization has tradeoffs — see Speakers → Local diarization.

---

Steps

1. Create an API key

Go to console.groq.com/keys and click Create API Key.

Key format: gsk_…

Keys are account-wide (not project-scoped).

Copy the key immediately — Groq shows it only once.

2. Configure transcription

Navigate to Quendian → Preferences → Transcription, pick Groq Whisper, and paste the key. Quendian uses whisper-large-v3 for transcription.

3. Configure summarization

Navigate to Preferences → Summarization, pick Groq, and paste the same key. Quendian uses llama-3.3-70b-versatile for summaries by default.

Same key, both roles

A single Groq API key authenticates both the transcription endpoint (whisper-large-v3) and the summarization endpoint (llama-3.3-70b-versatile). Quendian makes separate API calls per role.

---

Monitor usage and pricing

For current rates, see Groq’s pricing page. Track your usage at console.groq.com/settings/usage.