Pinned Tweet
#خاتم_النبیین_محمدﷺ
اللَّهُمَّ صل علَى مُحَمَّدٍ وعلَى آلِ مُحَمَّدٍ،كما صَلَّيْتَ علَى إبْرَاهِيمَ،وعلَى آلِ إبْرَاهِيمَ،إنَّكَ حَمِيدٌمَجِيدٌ
اللَّهُمَّ بَارِكْ علَى مُحَمَّدٍوعلَى آلِ مُحَمَّدٍ،كما بَارَكْتَ علَى إبْرَاهِيمَ،وعلَى آلِ إبْرَاهِيمَ إنَّكَ حَمِيدٌمَجِيدٌ
Zeenia retweeted
خان کی بہن کے بارے اسی طرح ہی رپورٹس ہیں جیسے خان کے بارے۔
کسی نے کچھ نہیں دیکھا بس وہ ہی سنا جو سنانے والوں نے سنانا چاہا۔
خان کیساتھ اب خان کی بہنوں کے بارے بھی نہ ملاقات ہوسکتی نہ خبر کی جاسکتی۔
پی ٹی آئی سہیل آفریدی کی جگہ پر کرنے کے چکر میں اور قوم کا بددعائیں دینے پر اظہار اطمینان ۔
Zeenia retweeted
For the record I love that I know more about football than my man, it’s one of my biggest flexes!
Zeenia retweeted
Scaling AI agents requires more than bigger models it requires smarter training systems
QwenGyre’s elastic GPU scheduling and trajectory processing show how 2 make long horizon agent training faster and more efficient
Better infrastructure, stronger agents, & less wasted compute
🤝 Paid partnership
Introducing QwenGyre, bringing elastic RL to extreme-long-horizon (XLong) agent training. ✨
It launches with two core designs: 🥳
- Elastic Scheduler: reallocates GPUs between rollout & training as demand shifts—live executions never interrupted. Up to 1.85× speedup over Colocate, 1.78× over Async.
- Trajectory Processor: turns branching executions (compaction, sub-agents, retries) into prefix-shared trajectory trees with bounded, deduplicated training samples. Partial progress after timeout still gets scored.
Scaled to Qwen3.8-2.4T-A95B with ~700K tokens per rollout, QwenGyre lifts NL2RepoBench from 52.5% to 58.5% in 48 steps, cutting end-to-end training time by ~44% vs Async.
🔗 Read the paper:
arxiv.org/abs/2609.33848
Zeenia retweeted
AI is transforming the insurance industry!
Combining cloud technology with AI-powered underwriting can help financial consultants make faster, smarter decisions and deliver a smoother customer experience.
A great example of AI turning innovation into practical business value!
🤝 Paid partnership
Developed through the collaboration between Prudential Hong Kong and Alibaba Cloud, AI Underwriter elevates customer experience while optimizing the insurance journey.
This solution streamlines the insurance process by enabling financial consultants to obtain preliminary underwriting indications instantly through a chatbot, driving practical innovation in the industry!
Zeenia retweeted
🎃 Halloween surprises just got more exciting! 👀🍬
Free AI generations, credit pack savings, or a special subscription discount? 🎁✨
Time to log in to @OpenArtAI and reveal your treat before November 1! 🚀
🤝 Paid partnership
Your Halloween treat is waiting on OpenArt! 🍬
This Halloween, we’re giving you a surprise reward.
Will you get:
🎁 A special discount on your subscription?
💳 Savings on credit packs?
✨ FREE generations on selected AI image and video models?
There’s only one way to find out! Log in to OpenArt and reveal your treat.
Ends November 1, 11:59 PM PT.
One random reward per eligible account. Offers vary. Free users must upgrade to redeem.