–
Not ranked
Claim this launch to manage it.
Fast AI inference for open models
Groq is an AI inference company that runs open models on its own specialized hardware and offers them to developers through a cloud API. It is built for teams that need fast, low-latency responses from large language models in production applications. Groq designs custom chips known as LPUs, which are built specifically for inference rather than training, and pairs them with a managed cloud platform covering infrastructure, model serving and usage controls. Developers can call popular open-weight models through an API and get high token throughput. The company has been expanding its data center capacity, and its newer LPX systems are designed to run alongside NVIDIA GPUs.
–
Not ranked
Claim this launch to manage it.
Sign in to join the conversation.
No comments yet. Share your feedback.