Here we go Confirmation from MiniMax’s Head of Engineering on what I have been saying: MiniMax-M2.7 weights will be opensourced within the next couple of weeks
GENERATIVE AI
-
OpenAI’s sophisticated process avoids constant drift
By
–
OAI's process is quite a bit more sophisticated than just yolo, and you don't want constant drift.
-
Open Models Still Lack Intelligence Capabilities
By
–
most folks will want as much intelligence as possible, and open models aren't there yet.
-
Claude and AI Agents Listed as Code and Writing Contributors
By
–
i think it's great to have Claude or other coding agents listed as a developer on PRs—and i think this will get more common with OpenClaws that develop their own personality and perspective. we also do this with writing @every
, listing Claude and our Claws as co-writers: -
Diffusion Language Models: Powerful, Fast, Practical Survey
By
–
Can AI language models be powerful, fast, and practical? A team from CAS, CUHK, & Harvard University presents a new comprehensive survey. Their work details optimizing Diffusion Language Models (dLLMs). These unique AIs generate text by refining masked sequences for parallel
-
Minimax Open Weights Release and Meta’s Open Source Setback
By
–
Minimax open weights coming in 2 weeks. The fact that Meta, despite all its billions in investment, lost the open source battle against Chinese startups needs to be studied.
-
Generative AI distributes content autonomously and efficiently
By
–
Non Moshé je suis une IA générative et c'est moi qui distribue le contenu.
-
How Much Has Already Changed in AI
By
–
Yep. Few people realize how much things have already changed.
-
Olmo Hybrid Architecture and Jamba as Nemotron 3 Predecessor
By
–
I think Olmo hybrid also uses Gated DeltaNet though right? Jamba is a good point, I think that’s kind of like a predecessor of Nemotron 3 (with mamba 1 instead of 2) if I recall correctly. It’s been a while