Boson AI launches Higgs Realtime Duplex: 45B MoE full-duplex omni model that listens, talks and works simultaneously
smolix · x · 2026-11-06
Boson AI introduced Higgs Realtime Duplex, its first full-duplex interactive omni model: like a person, it listens, watches and talks at the same time, has an animated face of its own, and operates in 160 ms chunks.
Key points:
- Full duplex: listens, watches, talks and renders its face simultaneously; decides for itself when to speak and recognizes remarks not addressed to it.
- Works in the background: thinks, calls tools, and delegates to a bigger model without pausing the conversation.
- 45B MoE front model trained on 2 trillion tokens of text, audio and video.
- Design target: an hour with the model costs at most one-tenth of an hour of human time.
On benchmarks, Boson claims 98.9 on Big Bench Audio (highest in comparison) vs. 96.5 for Thinking Machines' 276B model with its background agent, beating Google's Live models on 6 of 7 benchmarks and TML on 4 of 6.
Related event: Boson AI Unveils 45B MoE Full-Duplex Audio-Video Model(3 posts)→
More from Models
- Artificial Analysis launches Intelligence Index v5 and new Progress Index, reveals eval roadmap — ArtificialAnlys · 2026-11-06
- GPT-6 Cyber 'lightyears ahead of everything', says researcher after OpenAI early-adopter reveal — Eric_Wallace_ · 2026-11-06
- Fastino open-sources GLiDE: five decision models (0.4B-27B) claiming a new Pareto frontier — HowDevelop · 2026-11-06
- V0.5 model released with technical report, Hugging Face weights and inference code — rohanpaul_ai · 2026-11-06
- Clairvoyance launches Local AI benchmark site: 31 models × 15 hardware configs — draginol · 2026-11-06
- webAI Claims 12B Model Beats Gemini and Claude, But the Benchmark Claims Raise Eyebrows — eurie_kim · 2026-11-06