📆 Ce lundi 26 janvier 2026, à 15 heures (heure française) @oeg_francophone pose cette question impertinente et nécessaire: "L’Open Education a-t-elle encore un sens en 2026?" Venez découvrir les enjeux de l'accès universel à l'éducation, avec une présentation de notre organisation et le jeu développé avec Chaire Unesco RELIA "23 bonnes raisons", à l'heure de l'hashtag#IA et des "lumières sombres"

👀 Lancé en 2020, le Festival de l’Apprendre / Learning Planet Festival célèbre tous les apprentissages: la musique, la science, la philosophie, la couture... dans la rue, en classe ou en ligne, en toute liberté. En partenariat avec l’@unesco, il orchestre chaque année des centaines d’événements à travers le monde - 600 dans 193 pays en 2025 - autour de la célébration de la Journée internationale de l’éducation le 24 janvier.

Info évènement : https://www.linkedin.com/events/l-openeducationa-t-elleencoreun7419373814465216512/?lipi=urn%3Ali%3Apage%3Aorganization_admin_admin_page_posts_published%3Bc1bc2f12-150f-4189-8212-09aac14e74b7

#oegfr #oegf #openeducation #educationouverte #libreedu #communsnumeriques #openscience #openmodels

LinkedIn Login, Sign in | LinkedIn

Login to LinkedIn to keep in touch with people you know, share ideas, and build your career.

LinkedIn

Sujee Maniyam (@sujee_dev)

MiniMax의 최신 모델 M2.1이 @nebiustf에 공개되었습니다. M2 시리즈 대비 코딩 및 에이전트 워크플로 성능이 향상되었고, @ArtificialAnlys 기준 상위 오픈 모델로 평가받고 있다는 보고입니다. 실전 코딩·에이전트 활용에 주목되는 업데이트입니다.

https://x.com/sujee_dev/status/2011950073790513343

#minimax #m2.1 #openmodels #coding #agents

Sujee Maniyam (@sujee_dev) on X

MiniMax M2.1 from MiniMax is now live on @nebiustf A solid step up from the M2 series, M2.1 is showing strong results for **coding** and **agentic workflows**. It’s also ranked among the **top open models** by @ArtificialAnlys And yes… it passes my favorite vibe benchmark

X (formerly Twitter)

Artificial Analysis (@ArtificialAnlys)

GLM 4.7이 오픈 웨이트 모델들 가운데 Artificial Analysis Intelligence Index에서 선두를 차지했으며, 에이전트용(use cases)으로 가장 강력한 오픈 모델로 평가됩니다. 이로 인해 제공업체의 응답 속도와 가격 정책이 중요한 고려사항으로 부각된다는 개요입니다(출처: @Zai_org).

https://x.com/ArtificialAnlys/status/2011581689567592641

#glm4.7 #openmodels #agents #ai

Artificial Analysis (@ArtificialAnlys) on X

GLM 4.7 providers overview: GLM 4.7 leads the Artificial Analysis Intelligence Index among open weights models and is the strongest open weights model for agentic use cases - making provider speed and pricing critical GLM 4.7 from @Zai_org is now the most intelligent open

X (formerly Twitter)

J. Iwasawa (@jiwasawa)

Gemini 3 Pro가 5~10조(테라) 파라미터 규모일 것이라는 예측을 소개하는 트윗입니다. 해당 예측은 Artificial Analysis 창업자들이 오픈 모델의 AA-Omniscience Index와 총 파라미터 수의 관계를 외삽해 도출한 것이며, Gemini 3 Pro의 AA-Omniscience Index를 13으로 예상합니다.

https://x.com/jiwasawa/status/2010688287997137224

#gemini3pro #gemini #modelsize #aaomniscience #openmodels

J. Iwasawa (@jiwasawa) on X

Gemini 3 Pro は 5~10Tパラメータモデルと Artificial Analysis 創業者らが予想していて興味深い。 オープンモデルでの AA-Omniscience Index vs. Total Parameters の関係から外挿した予想。Gemini 3 Pro は AA-Omniscience Index が 13。

X (formerly Twitter)

Kimi.ai (@Kimi_Moonshot)

Jensen Huang이 "Open models really took off last year."라고 발언했다는 인용을 포함한 트윗으로, 작성자는 그 오픈 모델 확산 흐름에 함께하게 되어 기쁘다고 표현하고 있습니다. (오픈 모델의 확산이라는 AI 트렌드 언급)

https://x.com/Kimi_Moonshot/status/2009116264393003029

#openmodels #jensenhuang #aitrends

Kimi.ai (@Kimi_Moonshot) on X

Jensen Huang: "Open models really took off last year." We're so glad to be part of that.

X (formerly Twitter)

AI didn’t get expensive because it got smarter.
It got expensive because it never learned when to stop thinking. 🧠⚡

This article breaks down the open model that finally treats reasoning like a budgeted resource.

https://medium.com/@rogt.x1997/30b-parameters-3b-active-the-ai-model-that-cut-reasoning-costs-by-design-014808c57a39

#AgenticAI #LLMEconomics #OpenModels
https://medium.com/@rogt.x1997/30b-parameters-3b-active-the-ai-model-that-cut-reasoning-costs-by-design-014808c57a39

30B Parameters, 3B Active: The AI Model That Cut Reasoning Costs by Design

Why Nemotron 3 Nano changes the math of agentic AI

Medium
Hoe China de wereld van open AI modellen domineert, mooi overzicht. Zonder dat je het weet zit achter de toepassingen van veel bedrijven een chinees open model, al dan niet verbeterd/gefinetuned en dergelijke. https://www.interconnects.ai/p/8-plots-that-explain-the-state-of
#AI #china #openmodels #geopolitics
8 plots that explain the state of open models

Measuring the impact of Qwen, DeepSeek, Llama, GPT-OSS, Nemotron, and all of the new entrants to the ecosystem.

Interconnects
tiis-falcon-h1r-7b-can-out-reason-models-up-to-7x-its-size-and-its-mostly venturebeat.com/technology/tii… #AI #Falcon #OpenModels
NVIDIA Unveils New Open Models, Data & Tools to Accelerate AI

NVIDIA releases a suite of open models, massive multimodal datasets, and tools—from Nemotron speech to Alpamayo autonomous driving—empowering developers to build real‑world AI faster.

TechLife