Пополнение баланса
- Open models on LPU hardware with very high generation speed
- Pay-as-you-go token billing
- Speech transcription and an OpenAI-compatible endpoint
Home / APIs & tokens / Groq
Groq
Plan contents as published by the vendor; seller prices arrive at launch.
Groq runs open models on its own LPU accelerators and is known for extremely fast token generation. The catalog includes Llama, Qwen, GPT-OSS, Kimi and other models plus Whisper speech recognition, all behind an OpenAI-compatible API. It is a strong fit for voice assistants, agents and anything where response latency is user-visible.