Founder & CEO of @OnDemandAI
Abu Dhabi, UAE
Joined December 2021
- Tweets608
- Following510
- Followers9.7K
- Likes4.5K
MK retweeted
سيظل الوالد المؤسس الشيخ زايد بن سلطان آل نهيان، طيب الله ثراه، حاضراً في وجدان شعب الإمارات. قاد جهود تأسيس الاتحاد وأرسى دعائم الوطن على الوحدة والإخلاص والعطاء، وغرس في قلوب أبنائه قيماً أصيلة نعتز بها ونفخر، وترك إرثاً خالداً من الحكمة والبصيرة والرؤى الثاقبة. واليوم نعلن السابع من يناير من كل عام يوم زايد المؤسس مناسبة وطنية نستحضر فيها هذا الإرث الثري ونغرس مبادئه في نفوس الأجيال ليظل دائماً مصدر إلهام لطريقنا نحو المستقبل.
MK retweeted
I am standing to be Secretary-General because I believe this is a moment to act with greater urgency: to act earlier, bring people together across difference, and drive the work forward until it produces concrete results. That is how we rebuild trust in the United Nations and ensure it delivers meaningful change for the people it exists to serve.
My vision: lananusseibeh.org
Wow
🚀 Introducing DeepInfer, our in-house inference framework, now fully open source!
Powering DeepSeek-V4.1-Flash at scale with high throughput and low latency. We’re sharing our production stack to help the community lower compute costs, reduce energy use, and make AI more affordable.
🔗 github.com/deepseek-ai/DeepI…
1/4
😍
We trained our Muse foundational models, including Muse Wonder, at Prometheus, our gigawatt-scale supercluster in New Albany, Ohio.
Across a supercluster with hundreds of thousands of GPUs, even one underperforming host can slow an entire training run. We engineered Prometheus for resilience at scale, achieving 85%+ goodput on pretraining, 90%+ on reinforcement learning and ~47K GPU hours mean time between interruptions during Muse Wonder training.
Take a look at the tech powering Prometheus: go.meta.me/4yXXApZ
Been telling people this for the longest that’s why we built harness-arena.ai
Turns out DeepSeek harness improves the score of DeepSeek V4.1 Flash by 2-3 points on average.
It also costs 80% more due to subagents and more output tokens.
New results:
DeepSeek V4.1 Flash 1017 + Claude Code harness:
- max, mean of 7: 39.1
- high, mean of 5: 36.4
- low, mean of 5: 34.4
DeepSeek V4.1 Flash 1017 + DeepSeek harness:
- max, mean of 7: 42.7
- high, mean of 5: 39.2
- low, mean of 5: 34.8
For comparison:
GPT-6 Astra, max: 45
Opus 5.5, max: 41.7
Muse Wonder Full 1.0, max: 42
It's a crazy model, and the only one catched cheating. All the results are available on the benchmark website.
MK retweeted
@AIatMeta is humbly downplaying @Muse muse-wonder-full-1.0's coding ability in favor of its agentic assistant strengths.
Muse Code is Fable-level. Twice as fast. And significantly more generous usage limits.
I have been using it nonstop since its release and it has completely replaced Fable for my most challenging tasks. If there's one thing I'm good at, it's burdening agents with the most challenging of tasks.
MK retweeted
Introducing Muse Wonder, our new family of frontier models with state of the art performance for personal and professional agentic work.
MK retweeted
يمثل التوجه الاستثماري والتنموي والإنساني للإمارات تجاه القارة الأفريقية قصة نجاح متميزة. فقد أدركت الإمارات مبكراً أن العالم يتغير، وأن التواصل مع أفريقيا وبناء الشراكات معها لا ينبغي أن يمر عبر البوابة الأوروبية والغربية.
واليوم، يتجاوز حجم التبادل التجاري السنوي بين الإمارات والقارة ١٠٧ مليار دولار، فيما تقترب الاستثمارات الإماراتية فيها من ١٥٠ مليار دولار.
نجاح آخر لسياسة إماراتية تنوّع الشراكات وتعمّق التعاون وتبني جسوراً مباشرة مع عالم متغير.
MK retweeted
Grok 4.7 has appeared on Google cloud quotas.
This is usually a same-day release, generally up to 12 hours though.
MK retweeted
اطلعنا اليوم، خلال اجتماع المجلس الوزاري للذكاء الاصطناعي والتنمية، على المرحلة الثانية من مشروع "الذكاء الاصطناعي المساعد"، الذي يمثل خطوة جديدة في مسيرة تطوير حكومة المستقبل. ومستهدفنا واضح: تحويل إمكانات الذكاء الاصطناعي إلى أثر ملموس في كفاءة الحكومة، وجودة الخدمات، وسرعة الإنجاز. وكل مرحلة جديدة تقربنا من منظومة حكومية أكثر مرونة، تواكب متغيرات العصر، وتصمم خدماتها انطلاقاً من احتياجات الإنسان وتطلعاته، وتسخر التكنولوجيا في خدمته.
MK retweeted
Google is finally gearing up to release Gemini 4.
~2-3 days ago, [REDACTED] had 120+ new mentions of Gemini 4, up from 0. Less than 24 hours ago, [REDACTED] product got 6 new mentions similar to the other mentions.
It is starting to propagate.
MK retweeted
Google people HAVE to stop vagueposting about Ox Alpha.
If it's not their model, its gonna be the biggest embarassment of 2026.
I think @sama is just honest even if it goes against him which I respect .
Sam Altman repeated many times that the ideal model for him is a model that does not have a lot of world knowledge, is small by comparison and very cheap to run, and can reconstruct knowledge via reasoning as need arises
A paradox with that statement is that, if it were possible to construct such a model, then it also could be small enough to run locally. So why should people pay premium for your proprietary model, if they could have an open weight one running locally for free (*or at the price of electricity) (***there is actually a valid answer to this question, and that is speed, which OpenAI indeed started to focus on recently)
DS4 Flash and Qwen 27B line-up lately has been converging to this Sam Altman's platonic ideal of a model—the Altmanic model, if you will
Not rented proprietarily, but distributed and hacked on truly freely