@kyutai-labs.bsky.social open-sourced (model, dataset, code, and data pipeline) Helium 1, their new 2B-parameter LLM.
🧑💻 Blog: kyutai.org/2025/04/30/h...
🤗 Models on HuggingFace: huggingface.co/kyutai/heliu...
📚 GitHub: github.com/kyutai-labs/...
🧑💻 Blog: kyutai.org/2025/04/30/h...
🤗 Models on HuggingFace: huggingface.co/kyutai/heliu...
📚 GitHub: github.com/kyutai-labs/...
May 1, 2025 at 3:48 PM
@kyutai-labs.bsky.social open-sourced (model, dataset, code, and data pipeline) Helium 1, their new 2B-parameter LLM.
🧑💻 Blog: kyutai.org/2025/04/30/h...
🤗 Models on HuggingFace: huggingface.co/kyutai/heliu...
📚 GitHub: github.com/kyutai-labs/...
🧑💻 Blog: kyutai.org/2025/04/30/h...
🤗 Models on HuggingFace: huggingface.co/kyutai/heliu...
📚 GitHub: github.com/kyutai-labs/...
Kyutai 💜
#ventart #artmoots #artistsupport #knyoc #demonslayeroc #demonslayer #kny #kyutai #chibi #commissionsopen
#ventart #artmoots #artistsupport #knyoc #demonslayeroc #demonslayer #kny #kyutai #chibi #commissionsopen
March 29, 2025 at 10:28 AM
Today @kyutai-labs.bsky.social and @ellisinsttue.bsky.social launch @kesai.eu!
Robot learning is bottlenecked by the cost of physical interaction. Our mission is to advance the efficiency frontier of robust & safe physical AI through fully open and reproducible research.
kesai.eu/blog/2026-05...
Robot learning is bottlenecked by the cost of physical interaction. Our mission is to advance the efficiency frontier of robust & safe physical AI through fully open and reproducible research.
kesai.eu/blog/2026-05...
Kyutai and ELLIS Tübingen launch KE:SAI: A Premier Franco-German Partnership for Open Science in Physical AI
Kyutai and ELLIS Tübingen announce the official launch of KE:SAI – Kyutai ELLIS Scalable Autonomous Intelligence, a non-profit open science lab for physical AI.
kesai.eu
May 20, 2026 at 9:35 AM
Today @kyutai-labs.bsky.social and @ellisinsttue.bsky.social launch @kesai.eu!
Robot learning is bottlenecked by the cost of physical interaction. Our mission is to advance the efficiency frontier of robust & safe physical AI through fully open and reproducible research.
kesai.eu/blog/2026-05...
Robot learning is bottlenecked by the cost of physical interaction. Our mission is to advance the efficiency frontier of robust & safe physical AI through fully open and reproducible research.
kesai.eu/blog/2026-05...
Say hello to Kyutai STT: real-time streaming speech‑to‑text with lightning speed 🚀
Kyutai just open‑sourced STT models optimized for super low latency and high throughput. The 1B bilingual model (EN/FR) lags just 500 ms, while the 2.6B English‑only version hits sub‑2.5s delay.
Kyutai just open‑sourced STT models optimized for super low latency and high throughput. The 1B bilingual model (EN/FR) lags just 500 ms, while the 2.6B English‑only version hits sub‑2.5s delay.
June 19, 2025 at 2:14 PM
Say hello to Kyutai STT: real-time streaming speech‑to‑text with lightning speed 🚀
Kyutai just open‑sourced STT models optimized for super low latency and high throughput. The 1B bilingual model (EN/FR) lags just 500 ms, while the 2.6B English‑only version hits sub‑2.5s delay.
Kyutai just open‑sourced STT models optimized for super low latency and high throughput. The 1B bilingual model (EN/FR) lags just 500 ms, while the 2.6B English‑only version hits sub‑2.5s delay.
KYUTAI: KYUTAI RELEASES OVIE-FT-DL3DV, 0.1B PARAMS, ON HUGGING FACE
September 30, 2026 at 6:01 PM
KYUTAI: KYUTAI RELEASES OVIE-FT-DL3DV, 0.1B PARAMS, ON HUGGING FACE
Meet Helium-1 preview, our 2B multi-lingual LLM, targeting edge and mobile devices, released under a CC-BY license. Start building with it today!
huggingface.co/kyutai/heliu...
huggingface.co/kyutai/heliu...
kyutai/helium-1-preview-2b · Hugging Face
We’re on a journey to advance and democratize artificial intelligence through open source and open science.
huggingface.co
January 13, 2025 at 5:50 PM
Meet Helium-1 preview, our 2B multi-lingual LLM, targeting edge and mobile devices, released under a CC-BY license. Start building with it today!
huggingface.co/kyutai/heliu...
huggingface.co/kyutai/heliu...
KYUTAI: KYUTAI RELEASES OVIE-512, 41.6 MS PER VIEW ON H100
September 30, 2026 at 6:31 PM
KYUTAI: KYUTAI RELEASES OVIE-512, 41.6 MS PER VIEW ON H100
ANOTHER COMM!!
Again gorgeous darling Kyutai🩶 Thank you @minimarzzz.bsky.social for your support 🩶
#commissionedart #knyoc
Again gorgeous darling Kyutai🩶 Thank you @minimarzzz.bsky.social for your support 🩶
#commissionedart #knyoc
August 31, 2025 at 7:43 PM
ANOTHER COMM!!
Again gorgeous darling Kyutai🩶 Thank you @minimarzzz.bsky.social for your support 🩶
#commissionedart #knyoc
Again gorgeous darling Kyutai🩶 Thank you @minimarzzz.bsky.social for your support 🩶
#commissionedart #knyoc
Kyutai, le laboratoire IA de Xavier Niel, veut révolutionner la traduction vocale
Kyutai, le laboratoire IA de Xavier Niel, veut révolutionner la traduction vocale
Kyutai, le laboratoire IA de Xavier Niel, veut révolutionner la traduction vocale
www.numerama.com
February 6, 2025 at 5:09 PM
Kyutai, le laboratoire IA de Xavier Niel, veut révolutionner la traduction vocale
It's kind of wild how quick and easy it is to clone a voice with ~5 seconds of audio. github.com/kyutai-labs/...
uvx pocket-tts generate --voice "./my_sample.wav" --text "whatever you want it to say"
uvx pocket-tts generate --voice "./my_sample.wav" --text "whatever you want it to say"
GitHub - kyutai-labs/pocket-tts: A TTS that fits in your CPU (and pocket)
A TTS that fits in your CPU (and pocket). Contribute to kyutai-labs/pocket-tts development by creating an account on GitHub.
github.com
January 25, 2026 at 10:59 AM
It's kind of wild how quick and easy it is to clone a voice with ~5 seconds of audio. github.com/kyutai-labs/...
uvx pocket-tts generate --voice "./my_sample.wav" --text "whatever you want it to say"
uvx pocket-tts generate --voice "./my_sample.wav" --text "whatever you want it to say"
Neil Zeghidour has held talks with American and French investors to commercialize the voice AI tech developed at a Eric Schimdt-backed AI research lab Kyutai.
Former Meta And Google Researcher In Talks To Raise $60 Million For Voice AI Startup
Neil Zeghidour has held talks with American and French investors to commercialize the voice AI tech developed at a Eric Schimdt-backed AI research lab Kyutai.
www.forbes.com
October 9, 2025 at 12:40 PM
Neil Zeghidour has held talks with American and French investors to commercialize the voice AI tech developed at a Eric Schimdt-backed AI research lab Kyutai.
I tested Pocket TTS from Kyutai. It's fun to use for generating French sentences: the accent is strong, but it's still perfectly understandable, even though French doesn't seem to be in the training dataset!
Beyond that, the model is quite impressive especially knowing it has only 100M parameters
Beyond that, the model is quite impressive especially knowing it has only 100M parameters
GitHub - kyutai-labs/pocket-tts: A TTS that fits in your CPU (and pocket)
A TTS that fits in your CPU (and pocket). Contribute to kyutai-labs/pocket-tts development by creating an account on GitHub.
github.com
January 16, 2026 at 3:49 PM
I tested Pocket TTS from Kyutai. It's fun to use for generating French sentences: the accent is strong, but it's still perfectly understandable, even though French doesn't seem to be in the training dataset!
Beyond that, the model is quite impressive especially knowing it has only 100M parameters
Beyond that, the model is quite impressive especially knowing it has only 100M parameters
👻KYŪTAI👻
Hashira of Death
Artwork for @minimarzzz.bsky.social ! 💜
#kny #knyoc #ocart #demonslayer #kimetsunoyaiba
Hashira of Death
Artwork for @minimarzzz.bsky.social ! 💜
#kny #knyoc #ocart #demonslayer #kimetsunoyaiba
November 16, 2025 at 8:28 PM
👻KYŪTAI👻
Hashira of Death
Artwork for @minimarzzz.bsky.social ! 💜
#kny #knyoc #ocart #demonslayer #kimetsunoyaiba
Hashira of Death
Artwork for @minimarzzz.bsky.social ! 💜
#kny #knyoc #ocart #demonslayer #kimetsunoyaiba
Huggingface: huggingface.co/kyutai/heliu...
huggingface.co
January 15, 2025 at 1:37 AM
Huggingface: huggingface.co/kyutai/heliu...
Commission time!! Thank you so much @minimarzzz.bsky.social for your support and the honor to draw Kyutai. I love her.
#commission #art
#commission #art
August 24, 2025 at 9:40 PM
Commission time!! Thank you so much @minimarzzz.bsky.social for your support and the honor to draw Kyutai. I love her.
#commission #art
#commission #art
✦ I bled for things that never learned how to stay.
Bled my time, my patience, my love..
ft. Kyutai (Oc)
#artmoots #knyoc #commissionsopen #artist #demonslayeroc #demonslayer #kny #trippy #ventart
Bled my time, my patience, my love..
ft. Kyutai (Oc)
#artmoots #knyoc #commissionsopen #artist #demonslayeroc #demonslayer #kny #trippy #ventart
August 8, 2026 at 7:14 PM
✦ I bled for things that never learned how to stay.
Bled my time, my patience, my love..
ft. Kyutai (Oc)
#artmoots #knyoc #commissionsopen #artist #demonslayeroc #demonslayer #kny #trippy #ventart
Bled my time, my patience, my love..
ft. Kyutai (Oc)
#artmoots #knyoc #commissionsopen #artist #demonslayeroc #demonslayer #kny #trippy #ventart
“L’intelligence artificielle est fondamentalement une science, et si on veut que ça le reste, il faut que la communauté partage le plus possible ses trouvailles” www.ouest-france.fr/high-tech/in...
Sommet de l’IA : « Nous n’avons pas à rougir » souligne le PDG de Kyutai, cofondé par Xavier Niel
Patrick Pérez dirige Kyutai, un laboratoire à but non lucratif parisien qui a développé une intelligence artificielle vocale. À l’occasion du sommet pour l’action sur l’intelligence artificielle, orga...
www.ouest-france.fr
February 8, 2025 at 9:22 AM
“L’intelligence artificielle est fondamentalement une science, et si on veut que ça le reste, il faut que la communauté partage le plus possible ses trouvailles” www.ouest-france.fr/high-tech/in...
Hello world from @kesai.eu!
Here is a new non-profit open-science AI lab based in Tübingen and Paris pushing the frontiers of Physical AI and self-driving in particular.
They have a top-notch founding team, an ambitious agenda and incredible supporters (the team behind @kyutai-labs.bsky.social).
Here is a new non-profit open-science AI lab based in Tübingen and Paris pushing the frontiers of Physical AI and self-driving in particular.
They have a top-notch founding team, an ambitious agenda and incredible supporters (the team behind @kyutai-labs.bsky.social).
Open-source fueled the LLM revolution, but Physical AI hasn't fully benefited from this flywheel yet. Today, we're launching kesai.eu, our mission to democratize robotics research! First milestone: training a frontier-level self-driving policy using significantly less data than typically required.
KE:SAI — Open Science Autonomy Lab
KE:SAI is a Franco-German non-profit open science lab for scalable autonomous intelligence.
kesai.eu
May 20, 2026 at 8:22 AM
Hello world from @kesai.eu!
Here is a new non-profit open-science AI lab based in Tübingen and Paris pushing the frontiers of Physical AI and self-driving in particular.
They have a top-notch founding team, an ambitious agenda and incredible supporters (the team behind @kyutai-labs.bsky.social).
Here is a new non-profit open-science AI lab based in Tübingen and Paris pushing the frontiers of Physical AI and self-driving in particular.
They have a top-notch founding team, an ambitious agenda and incredible supporters (the team behind @kyutai-labs.bsky.social).
Also there is there is also Kyutai with:
- Moshi speech-text model github.com/kyutai-labs/...
- MoshiVis: vision-speech-text model github.com/kyutai-labs/...
- Moshi speech-text model github.com/kyutai-labs/...
- MoshiVis: vision-speech-text model github.com/kyutai-labs/...
GitHub - kyutai-labs/moshi: Moshi is a speech-text foundation model and full-duplex spoken dialogue framework. It uses Mimi, a state-of-the-art streaming neural audio codec.
Moshi is a speech-text foundation model and full-duplex spoken dialogue framework. It uses Mimi, a state-of-the-art streaming neural audio codec. - kyutai-labs/moshi
github.com
August 18, 2025 at 2:00 PM
Also there is there is also Kyutai with:
- Moshi speech-text model github.com/kyutai-labs/...
- MoshiVis: vision-speech-text model github.com/kyutai-labs/...
- Moshi speech-text model github.com/kyutai-labs/...
- MoshiVis: vision-speech-text model github.com/kyutai-labs/...
🎉 Another invited speaker announcement for TokShop!
Welcome @honualx.bsky.social , co-founder of @kyutai-labs.bsky.social, speaking on:
"Discrete and continuous tokenization for audio-text modeling" 🧵
Welcome @honualx.bsky.social , co-founder of @kyutai-labs.bsky.social, speaking on:
"Discrete and continuous tokenization for audio-text modeling" 🧵
September 29, 2026 at 8:20 AM
🎉 Another invited speaker announcement for TokShop!
Welcome @honualx.bsky.social , co-founder of @kyutai-labs.bsky.social, speaking on:
"Discrete and continuous tokenization for audio-text modeling" 🧵
Welcome @honualx.bsky.social , co-founder of @kyutai-labs.bsky.social, speaking on:
"Discrete and continuous tokenization for audio-text modeling" 🧵
We just released the Helium-1 model , a 2B multi-lingual LLM which @exgrv.bsky.social and @lmazare.bsky.social have been crafting for us! Best model so far under 2.17B params on multi-lingual benchmarks 🇬🇧🇮🇹🇪🇸🇵🇹🇫🇷🇩🇪
On HF, under CC-BY licence: huggingface.co/kyutai/heliu...
On HF, under CC-BY licence: huggingface.co/kyutai/heliu...
January 13, 2025 at 6:10 PM
We just released the Helium-1 model , a 2B multi-lingual LLM which @exgrv.bsky.social and @lmazare.bsky.social have been crafting for us! Best model so far under 2.17B params on multi-lingual benchmarks 🇬🇧🇮🇹🇪🇸🇵🇹🇫🇷🇩🇪
On HF, under CC-BY licence: huggingface.co/kyutai/heliu...
On HF, under CC-BY licence: huggingface.co/kyutai/heliu...
Gradium, a startup spawned out of French AI lab Kyutai (backed by French telecom billionaire Xavier Niel), launched out of stealth on Tuesday.
Paris-based AI voice startup Gradium nabs $70M seed | TechCrunch
Gradium, a startup spawned out of French AI lab Kyutai (backed by French telecom billionaire Xavier Niel), launched out of stealth on Tuesday.
techcrunch.com
December 2, 2025 at 12:05 PM
Gradium, a startup spawned out of French AI lab Kyutai (backed by French telecom billionaire Xavier Niel), launched out of stealth on Tuesday.
hey so this is really cool !!! github.com/kyutai-labs/...
im running this on my old as shit thinkpad in my agents env and its made our STT<->TTS loop so much faster :3 it's a 100M param model (!)
our integral pieces: moonshine STT, pocket-tts with a voice clone, and ofc the Letta Agent SDK :D
im running this on my old as shit thinkpad in my agents env and its made our STT<->TTS loop so much faster :3 it's a 100M param model (!)
our integral pieces: moonshine STT, pocket-tts with a voice clone, and ofc the Letta Agent SDK :D
GitHub - kyutai-labs/pocket-tts: A TTS that fits in your CPU (and pocket)
A TTS that fits in your CPU (and pocket). Contribute to kyutai-labs/pocket-tts development by creating an account on GitHub.
github.com
August 27, 2026 at 4:11 AM
hey so this is really cool !!! github.com/kyutai-labs/...
im running this on my old as shit thinkpad in my agents env and its made our STT<->TTS loop so much faster :3 it's a 100M param model (!)
our integral pieces: moonshine STT, pocket-tts with a voice clone, and ofc the Letta Agent SDK :D
im running this on my old as shit thinkpad in my agents env and its made our STT<->TTS loop so much faster :3 it's a 100M param model (!)
our integral pieces: moonshine STT, pocket-tts with a voice clone, and ofc the Letta Agent SDK :D
Congrats and have fun at Kyutai!
July 3, 2025 at 6:43 AM
Congrats and have fun at Kyutai!
Ce matin j’ai testé en direct sur @franceculture.fr l’agent vocal « unmute » développé par le labo Kyutai (Niel,Saade, Schmidt), il simule à merveille la discussions en jouant sur les silences et hésitations.
Expérience à écouter ici 👇
Expérience à écouter ici 👇
May 29, 2025 at 7:34 AM
Ce matin j’ai testé en direct sur @franceculture.fr l’agent vocal « unmute » développé par le labo Kyutai (Niel,Saade, Schmidt), il simule à merveille la discussions en jouant sur les silences et hésitations.
Expérience à écouter ici 👇
Expérience à écouter ici 👇