Raon-SpeechChat becomes first open-source full-duplex speech model in top quadrant on Artificial Analysis
Kangwook_Lee · x · 2026-09-22
Raon-SpeechChat has landed on Artificial Analysis as the first open-source full-duplex speech model in the 'Most Attractive Quadrant': #1 in Speech Reasoning (57.5%) and #2 in Conversational Dynamics (86.2%), with model, code, and benchmark all public. The team also highlighted A.X K2 Raon-Speech (SK Telecom's Korean Sovereign AI project, #1 Korean/#3 English) and Raon-Speech-9B.
More from Models
- OpenRouter silently ignores effort params — benchmark via native API endpoints, not assumptions — PawelHuryn · 2026-09-22
- Abliterated Qwen3.6-35B-A3B released in MLX format for local Mac inference — evilsocket · 2026-09-22
- Probes confirm Xiaomi's MiMo-V2.6-Pro reasoning effort dial does nothing on OpenRouter — PawelHuryn · 2026-09-22
- Transformers Now Runs GGUF at llama.cpp Speed via GGML Kernels — LysandreJik · 2026-09-22
- 68-Second OpenAI Agents API Session Racked Up $132 in 'Container Usage' — 366 Hours of Billed Time With No Trace — Emotional-Orange- · 2026-09-22
- ChatGPT Free Tier Asked a Coder to Paste Terminal Output 30-40 Times Before Burning Their Message Limit — Active-Ad9741 · 2026-09-22