Finds. A 25829PT, 1 of 156 examples made in platinum. This is the second generation skeletonized AP RO QP, featuring feuille hands, leap indication, and updated typogrpahy. One of the most arresting Royal Oak QPs made, these are highly overlooked in light of the 25636. https://t.co/koh7Ts4VoU
Alibaba Qwen has just released a non-thinking model even more powerful than Kimi K2...
And even better than Claude Opus 4 🤯
→ 100% open source
→ Only 22B active parameters
→ Available for free in Qwen Chat
All the links below
Today we are releasing ether0, our first scientific reasoning model.
We trained Mistral 24B with RL on several molecular design tasks in chemistry. Remarkably, we found that LLMs can learn some scientific tasks more much data-efficiently than specialized models trained from scratch on the same data, and can greatly outperform frontier models and humans on those tasks. For at least a subset of scientific classification, regression, and generation problems, post-training LLMs may provide a much more data-efficient approach than traditional machine learning approaches. 1/n
Releasing INTELLECT-2: We’re open-sourcing the first 32B parameter model trained via globally distributed reinforcement learning:
• Detailed Technical Report
• INTELLECT-2 model checkpoint
https://t.co/tIKbtUlJQH
Bitsandbytes latest works with `torch.compile(fullgraph=True)` and you should put it to good use 🔥
For example, when applied to Flux, it beefs up the performance quite a bit.
Code:
https://t.co/PyanMr1oPy
Enjoy 🔥
Cztery nowe satelity SAR dodane do konstelacji! Połączyliśmy się już ze wszystkimi oraz z sukcesem zakończyliśmy LEOP (launch and early operation). To już prawie 50! Po kalibracji satelity zostaną skierowane do wsparcia działań na Ukrainie.
https://t.co/FujYjR4bRr
Sonnet being rank 8 in instruction following and code is confusing the shit out of me because none of this matches my real world experience using these models. I feel like I’m either dumb or these rankings are completely inflated
🚨Chamath Palihapitiya on Google's Willow chips cracking Bitcoin: "When Sundar posted this I missed my next meeting because I had to figure this out. We need about 8000 chips to break SHA-256. We're in the 2-5 year shot clock for chains reimplementing new hashing algorithms."