Subscribe
Models
Download on Hugging Face
69 of 204
SaveFull-duplexWeightsCommercialLanguages
Shizuku-DuplexShizuku AI
Full-duplex2026-11——Unclearja
Haru-4Bhigurashi
Full-duplex2026-11—openYesen es pt +6
EgoVoice Paper onlyUniversity of Seoul
Full-duplex2026-10——Not releaseden
Full-duplex2026-10396 ms—Not releaseden
Full-duplex2026-10——Not releaseden
SteerablePlex Paper onlyUniversity of Illinois Urbana-Champaign et al.
Full-duplex2026-10——Not released—
Full-duplex2026-10——Yes—
Full-duplex2026-10——Yesen es fr +17
Phonic-1Phonic
Full-duplex2026-10——Yes51
Full-duplex2026-10——Unclear—
Realtime-VenusAnt Group (inclusionAI)
Full-duplex2026-09—openYesen zh
GanderOmni-Interaction-Gander
Full-duplex2026-09—openYes—
Full-duplex2026-09——Not releasedzh en
CharDuplex Paper onlyNanyang Technological University / AI Singapore / UC Berkeley
Full-duplex2026-09——Not released—
RePlay Paper onlyDisney Research Imagineering / BUT Speech@FIT
Full-duplex2026-09——Not released—
SALMONN-duo Paper onlyTsinghua University / ByteDance
Full-duplex2026-09——Not releaseden
SteerDuplex Paper onlyScale AI / University of Maryland
Full-duplex2026-09——Not releaseden
Gemini 3.8 Live APIGoogle DeepMind
Full-duplex2026-09——Yesen ja zh +94
Full-duplex2026-09——Unclear—
Full-duplex2026-09——Yes113
Full-duplex2026-09——Yes60
Full-duplex2026-09——Yes97
Full-duplex2026-08448 msopenYesen
SeedRealtimeByteDance Seed
Full-duplex2026-08——Unclear—
Full-duplex2026-08——Yeszh en ja +11
Full-duplex2026-08700 ms—Yes100+
DuplexOmniXi'an Jiaotong University / Peking University / Meituan
Full-duplex2026-07506 msopenYeszh en
Lychee-FDHITsz-TMG
Full-duplex2026-07—openYeszh en
Full-duplex2026-07——Yesen ja zh +52
Full-duplex2026-07798 ms—Yes—
BayLing-DuplexICT, Chinese Academy of Sciences
Full-duplex2026-06—openYes—
DyaPlex Paper onlyNVIDIA / HKUST
Full-duplex2026-06——Not released—
Moshi-Face Paper onlyNagoya University
Full-duplex2026-06——Not released—
Wan-Streamer Paper onlyAlibaba Group (Wan Team)
Full-duplex2026-06——Not releasedzh en
TML-Interaction-SmallThinking Machines Lab
Full-duplex2026-05400 ms—Unclearen
CounselPlexKAIST (Integrated Vision and Language Lab)
Full-duplex2026-05—openUnclearen
DuplexSLA Paper onlyStepFun / Peking University / Nanyang Technological University
Full-duplex2026-05——Not released—
HydraSmallest AI
Full-duplex2026-05864 ms—Yesen
Full-duplex2026-05——Yesen ja zh +67
Full-duplex2026-04—openNon-commercialen
ELLSATsinghua University / ByteDance
Full-duplex2026-04—openUnclearen
KAMESakana AI
Full-duplex2026-04—openUnclearen
Full-duplex2026-04——Yeszh en
Full-duplex2026-03—openResearch onlyzh en
Human-1Josh Talks
Full-duplex2026-032021 msopenYeshi
FLAIR Paper onlyNanyang Technological University / Mila
Full-duplex2026-03——Not released—
Full-duplex2026-03——Yesen ja zh +21
LLM-jp-Moshi-v1NII (LLM-jp)
Full-duplex2026-02—openYesja
MiniCPM-o 4.5OpenBMB / ModelBest
Full-duplex2026-02—openYesen zh +30
PersonaPlexNVIDIA (ADLR)
Full-duplex2026-01170 msopenYesen
Gemini 2.5 LiveGoogle DeepMind
Full-duplex2025-12——Yesen ja zh +21
Full-duplex2025-12——Yesen es fr +4
SCoT (Streaming Chain-of-Thought) Paper onlyCarnegie Mellon University / Sony Group
Full-duplex2025-10——Not released—
Qwen3-OmniAlibaba Cloud
Full-duplex2025-09—openYesen zh ja +16
Full-duplex2025-08—openYeszh en
FLM-AudioCofeAI (BAAI)
Full-duplex2025-07—openResearch onlyzh en
Kimi-AudioMoonshot AI
Full-duplex2025-04—openYeszh en
Full-duplex2025-04—openUnclearen zh fr +3
Sesame CSMSesame AI
Full-duplex2025-03—openYesen
HibikiKyutai
Full-duplex2025-02—openYesfr en
J-Moshi / J-Moshi-extNagoya University (Higashinaka Lab)
Full-duplex2025-01—openNon-commercialja
SALMONN-omni Paper onlyByteDance · Tsinghua · Cambridge
Full-duplex2024-11——Not releaseden
Hertz-devStandard Intelligence
Full-duplex2024-11120 msopenYes—
GLM-4-VoiceZhipu AI (Z.ai)
Full-duplex2024-10—openUnclearzh en
OmniFlatten Paper onlyAlibaba (Tongyi Lab)
Full-duplex2024-10——Not releasedzh en
MoshiKyutai
Full-duplex2024-09160 msopenYesen
SyncLLM (Synchronous LLMs) Paper onlyMeta AI / University of Washington
Full-duplex2024-09——Not released—
LSLM (Listening-while-Speaking Language Model) Paper onlyShanghai Jiao Tong University (X-LANCE) / ByteDance
Full-duplex2024-08——Not releaseden
dGSLMMeta AI
Full-duplex2023-01—openUnclearen