sui (@birdabo)
이번 주는 LLM 분야에서 가속화가 예상되며, 여러 새로운 모델들이 잇달아 공개될 것이라고 언급했다. 구체적 모델명은 없지만 AI 모델 출시가 집중되는 흐름을 예고하는 내용이다.
sui (@birdabo)
이번 주는 LLM 분야에서 가속화가 예상되며, 여러 새로운 모델들이 잇달아 공개될 것이라고 언급했다. 구체적 모델명은 없지만 AI 모델 출시가 집중되는 흐름을 예고하는 내용이다.
Tencent HY (@TencentHunyuan)
HY-Embodied-0.5가 공개되었으며, 실제 세계의 임바디드 에이전트를 위한 파운데이션 모델 패밀리입니다. 2B 모델은 오픈소스로 공개됐고, 공간-시간 인지와 임바디드 추론을 강화해 예측, 상호작용, 계획 성능을 높입니다.
https://x.com/TencentHunyuan/status/2042503238877135336
#foundationmodels #opensource #embodiedai #robotics #aimodel

We're releasing HY-Embodied-0.5, a family of foundation models for real-world embodied agents. The 2B model is now open source. It strengthens spatial-temporal perception and embodied reasoning for prediction, interaction, and planning. 🤖 The suite includes: 🔹 2B for edge
MiniMax (official) (@MiniMax_AI)
HumanX 2026에서 MiniMax의 Featured Talk가 진행됐으며, Linda Sheng가 Eric Newcomer와 함께 MiniMax가 대규모로 범용 지능을 구축하는 방식과 foundation model, 신뢰, 팀, 그리고 실제 제품이 무엇인지에 대해 공유했다.

Wrapped our Featured Talk at #HumanX2026. 🎙️ Linda Sheng, our President of Global Business, sat down with @EricNewcomer and shared how MiniMax builds general intelligence at scale, from foundation models to trust, team, and what the real product actually is. Key quotes from
Generalist AI veröffentlicht das multimodale Robotik-Modell GEN-1 mit 99 Prozent Erfolgsquote bei physischen Aufgaben.
Die Architektur verzichtet auf Teleoperation im Pre-Training und nutzt 500.000 Stunden Sensor-Daten menschlicher Hände für spontane Echtzeit-Fehlerkorrektur. Eine Adaption an neue Hardware benötigt nur eine Stunde spezifischer Roboterdaten.
#GeneralistAI #Robotik #FoundationModels #LLM #News
https://www.all-ai.de/news/news26/gen-1-generalist-new
[Перевод] Исследование макросов @Generable и @Guide во фреймворке FoundationModels
Приветствую, Хабр! Макросы Swift предоставляют мощный механизм для генерации кода, позволяя разработчикам уменьшать количество шаблонного кода и повышать читаемость. Фреймворк FoundationModels представляет новые макросы, призванные упростить генерацию данных для определённых типов моделей с использованием языковых моделей.
https://habr.com/ru/articles/1019392/
#Generable #guide #swift #swiftui #FoundationModels #swift_ai #swiftui_ai
Lukas Ziegler (@lukas_m_ziegler)
Franka Robotics와 Agile Robots SE 공동 부스에서 로봇과 파운데이션 모델이 실제로 결합되는 데모가 소개됐다. FR3 Duo를 포함한 시스템을 통해 연구용 로봇 하드웨어와 AI 기반 제어/인지가 어떻게 만나고 있는지 보여주는 현장 사례다.
https://x.com/lukas_m_ziegler/status/2037122390367326580
#robotics #foundationmodels #franka #agilerobots #demonstration

Where robotics + foundation models are actually meeting! 🔬 I came across a really interesting demo at the joint booth of @FRANKAROBOTICS and Agile Robots SE, and it’s a good snapshot of where things are heading. The first system is the FR3 Duo, basically two Franka Research 3
Meta stellt mit TRIBE v2 ein Foundation-Modell zur Vorhersage menschlicher Gehirnaktivitäten vor. Das System verarbeitet visuelle, auditive und linguistische Daten in einem gemeinsamen Vektorraum, trainiert auf über 500 Stunden fMRI-Daten. Bei der Auswertung von Filmreizen erzielt das Modell einen Vorhersagewert von 0,28, deutlich vor etablierten Analysemethoden.
#Meta #TRIBEv2 #Neuroscience #FoundationModels #News
https://www.all-ai.de/news/news26/meta-tribe-v2-neu
We are hiring! The Institute for Artificial Intelligence at the University of Stuttgart (@Uni_Stuttgart) is looking for a PostDoc to work on Foundation Models for Knowledge Graphs.
The position focuses on automating the management of data and knowledge by combining machine learning, logics, and natural language understanding. You will define new projects, mentor students, and contribute to real-world applications.
Full-time (100% TV-L E13), 2 years. Application deadline: April 16, 2026.
Apply via careers.uni-stuttgart.de — details here:
https://www.ki.uni-stuttgart.de/institute/news/PostDoc-working-on-Foundation-Models-for-Knowledge-Graphs-for-2-years-100-TV-L-E13/
#Hiring #PostDoc #KnowledgeGraphs #FoundationModels #AI #NLP #MachineLearning #SemanticWeb #AcademicJobs
FoundationModels' 4,096-token context window can be tough to manage. iOS 26.4 introduces new APIs to measure instructions, prompts, and full transcripts - finally giving us proactive token management instead of just reacting to errors when the limit is hit.
🔗: https://artemnovichkov.com/blog/tracking-token-usage-in-foundation-models by Artem Novichkov