← All repos

gpt-omni/mini-omni

View on GitHub ↗ ★ 3,585 Python

open-source multimodal large language model that can hear, talk while thinking. Featuring real-time end-to-end speech input and streaming audio output conversational capabilities.

3 mentions on Hacker News by 3 people: 1 story, 2 comments · first 2024-09-03 · latest 2024-11-04