@kyutai_labs: We're releasing Voice of Reason, a speech-native model that does math out loud. Give it a spoken problem without transc…

We’re releasing Voice of Reason, a speech-native model that does math out loud. Give it a spoken problem without transcription nor text LLM in the loop, and it reasons and answers in speech. GSM8K goes from 27.3% for GLM-4-Voice to 77.1%. Link in 🧵

Math is a good benchmark for models’ intelligence but speech models typically lag behind text ones. Speech models have to answer out loud immediately, but cascading around a text model (transcribe, think, speak) adds too much latency. How do we bring them up to par ?

We start from GLM-4-Voice, finetune it on synthesized spoken question answering, then post-train it with RL against a verifiable reward. The model answers straight-out, with no extra reasoning tokens at all, and reaches 70.3% accuracy, above some past speech models that needed explicit reasoning traces.

We can further enable thinking without speaking using the streaming reasoning scheme of STITCH (Chiang et al., https://arxiv.org/abs/2507.15375). The model reasons in silent chunks while its previous sentence is still playing: 77.1% on GSM8K, without added thinking latency.

Paper: https://arxiv.org/abs/2609.18677 (COLM 2026) Weights: https://hf.co/kyutai/glm-4-voice-of-reason-9b…, directly answers. https://hf.co/kyutai/glm-4-voice-of-reason-stitch-9b…, for extra reasoning tokens while speaking.