Inference news

Google has introduced Gemma 4, a new version of its open model. It uses multi-token prediction to generate tokens up to three times faster without sacrificing quality. This major performance boost can significantly reduce inference costs and improve user experience for developers and businesses.

AI chipmaker Groq is reportedly raising $650 million in an internal funding round. The move signals a strategic shift for the company, which plans to pivot from hardware sales to concentrate on its AI inference services, aiming to improve the speed and efficiency of AI model responses.