Kyutai has released Voice of Reason, 2 open-weight speech-to-speech models built on GLM-4-Voice-9B. Supervised fine-tuning and reinforcement learning lift spoken GSM8K accuracy from 27.3% to 77.1%. There is no transcription step and no text LLM in the loop. Both checkpoints are on Hugging Face and run on a single H100. The post Kyutai Releases Voice of Reason: A Speech-Native Model that Solves Spoken Math with Reinforcement Learning appeared first on MarkTechPost.
Read the original at MarkTechPost: Kyutai Releases Voice of Reason: A Speech-Native Model that Solves Spoken Math with Reinforcement Learning
