Blog post: mistral.ai/news/voxtral-tts Playground: console.mistral.ai/build/audio/text-to-speech Technical report: mistral.ai/static/research/voxtral-tts.pdf Model weights: huggingface.co/mistralai/Voxtral-4B-TTS-2603 [Translated from EN to English]
Our first speech model, Voxtral TTS, is out. It delivers SOTA performance while significantly reducing cost compared to existing solutions, and it operates with very low latency. It uses a new architecture that combines auto-regressive generation of semantic speech tokens with flow-matching for acoustic tokens. We are also releasing a technical report sharing all our training methodology and insights. Much more to come in audio — stay tuned ! Mistral AI (@MistralAI) 🔊Introducing Voxtral TTS: our new frontier open-weight model for natural, expressive, and ultra-fast text-to-speech 🎭Realistic, emotionally expressive speech. 🌍Supports 9 languages and accurately captures diverse dialects. ⚡Very low latency for time-to-first-audio. 🔄Easily adaptable to new voices — https://nitter.net/MistralAI/status/2037183026539483288#m
There’s a lot of activity around robotics, for sure. None of it points to an inflection. No, even Tesla can’t bring us closer to humanoid robots Hyperbole about humanoid robots being on the cusp of commercialization continues to build. The best research on the matter continues to show otherwise: the most basic technological requirements have yet to be solved, and there is little evidence they will be anytime soon. thetechnologyletter.com $TSLA $WATT $PDYN $SYM $NVDA
Great work by Chroma training a search agent with SoTA efficiency. Lots of cool details: a prune tool for editing context mid-search, a synthetic data pipeline with verification steps, and a curriculum that shifts from recall to precision. Trained with Tinker! https://t.co/csp0FoXCtH
Great work by Chroma training a search agent with SoTA efficiency. Lots of cool details: a prune tool for editing context mid-search, a synthetic data pipeline with verification steps, and a curriculum that shifts from recall to precision. Trained with Tinker! Chroma (@trychroma) Introducing Chroma Context-1, a 20B parameter search agent. > pushes the pareto frontier of agentic search > order of magnitude faster > order of magnitude cheaper > Apache 2.0, open-source — https://nitter.net/trychroma/status/2037243681988894950#m
How do we ensure AI in healthcare benefits everyone? From accelerating discovery to supporting clinicians and improving patient outcomes, explore what's real, what's next, and what's needed at the 2026 #RAISEHealth Symposium on June 2: @StanfordMed med.stanford.edu/raisehealth…
So excited that our work is on the cover of Science!!! We find that AI models overly affirm users, even when they describe harmful actions. Advice from sycophantic AI made people more self-centered, yet people prefer and trust it more, which may promote this model behavior. Myra Cheng (@chengmyra1) AI always calling your ideas “fantastic” can feel inauthentic, but what are sycophancy’s deeper harms? We find that in the common use case of seeking AI advice on interpersonal situations—specifically conflicts—sycophancy makes people feel more right & less willing to apologize. — https://nitter.net/chengmyra1/status/1974244768919282135#m
Maybe my Build A Large Language Model From Scratch book would be a good start for the foundation: https://
amzn.to/4fqvn0D
Followed by my LLM architecture comparison article: https://
magazine.sebastianraschka.com/p/the-big-llm-
architecture-comparison
…