Alibaba-affiliate Ant uses Chinese, U.S. chips to cut AI costs https://www.cnbc.com/2025/03/24/alibaba-affiliate-ant-uses-china-us-chips-to-cut-ai-costs.html
“Today’s the day when art scales with compute. Soon Hollywood will budget not for hours of hand labor, but for hours of compute running and re-running inference. Artists create “raw art” in prompts (sketches, color palette, style elements) and let machines interpolate the rest.” / X https://x.com/DrJimFan/status/1904983307206713551
U.S. blacklists over 50 Chinese companies in bid to curb Beijing’s AI, chip capabilities https://www.cnbc.com/2025/03/26/us-blacklists-50-chinese-companies-in-bid-to-curb-beijings-ai-chip-capabilities.html
“Nvidia has scooped up inference provider Lepton AI in a deal worth several hundred million dollars. It’s Nvidia’s latest deal that’ll help it beef up its software offerings and make it easier for customers to use its GPUs. W/ @KevKubernetes: https://x.com/steph_palazzolo/status/1904947599368499497
“1X CEO Bernt Børnich told TechCrunch that by the end of 2025, NEO Gamma will be in “a few hundred to a few thousand” homes for early testing. “We want it to live and learn among people,” he stated. https://x.com/TheHumanoidHub/status/1904587022356144512
“Not AI generated https://x.com/TheHumanoidHub/status/1903175867846758840
“This ain’t CGI—but RTX mode is definitely ON! Jensen Huang gets a custom leather jacket hand-delivered by NEO, courtesy the brilliant minds at 1X. Literally can’t wait for y’all to see what’s coming at TED2025. It’s gonna be legendary. https://x.com/bilawalsidhu/status/1903107700415779196
“The Future of AI Agents at Nvidia” / X https://x.com/AtomSilverman/status/1902472452716052633
“Our ongoing and deep partnership with @NVIDIA spans the early days of Android to today’s cutting-edge AI collaborations across Alphabet. I’m really excited about the next phase of our partnership as we work together on agentic AI, robotics, and bringing the benefits of AI to more” / X https://x.com/sundarpichai/status/1902110575871848838
“@GroqInc API docs are now completely LLM-friendly with /llms.txt and /llms-full.txt files. This is the future of technical docs – easily accessible knowledge for models, not just humans. Add these to your model’s context window and let your AI agents build fast with our API. https://x.com/ozenhati/status/1902544634825330724
“NVIDIA announced major agentic AI initiatives at GTC 2025, focusing on infrastructure, models, and development tools & is working with Google & Oracle on agents. I summarized everything you need to know. 🧵 https://x.com/AtomSilverman/status/1902087634492780764
“We got lots of great community feedback on our open-source GR00T N1! Check out our Github, star, fork, contribute back! Let’s solve generally intelligent robots together, one commit at a time. https://x.com/DrJimFan/status/1902767546438148345
“.@NVIDIA built GR00T N1, a powerful open-source AI model designed for humanoid robots. It works across many types of robots — from robot arms (like LeRobot) to full humanoids, such as GR1 and 1X Neo robots. What is GROOT N1 at its core? ▪️ It’s a Vision-Language-Action (VLA) https://x.com/TheTuringPost/status/1903066519128641809
A key DeepMind robotics researcher left Google, and Nvidia has already backed his stealth startup | TechCrunch https://techcrunch.com/2025/03/19/a-key-deepmind-robotics-researcher-left-google-and-nvidia-has-already-backed-his-stealth-startup/
“This expressive robot designed by Disney is brought to life using the new open-source physics simulator accelerated with NVIDIA’s software framework. https://x.com/TheHumanoidHub/status/1902621870295543936
[2501.16295v1] Mixture-of-Mamba: Enhancing Multi-Modal State-Space Models with Modality-Aware Sparsity https://arxiv.org/abs/2501.16295v1
“TIL: You can deploy @GoogleDeepMind Gemma 3 with just 3 lines of code to @googlecloud Vertex AI! 👀 The new Model Garden SDK abstracts away the underlying infrastructure setup, making open model deployment incredibly simple. Learn more ⬇️ https://x.com/_philschmid/status/1905149450059727226
“The second Nvidia book to come out this year, complementing The Nvidia Way with more of a focus on Jensen personally. I enjoyed it, but I am slightly miffed about having a quote from me fabricated: “We wanted a faster responding Quake so our customers could reach out and frag https://x.com/ID_AA_Carmack/status/1904958211767034205
“NVIDIA released new Mamba image backbones on @huggingface 🔥 five models in sizes 97M, 230M and 740M, supporting resolutions 224, 256 and 512 👀 easily load with transformers 🤗 https://x.com/mervenoyann/status/1904168637612630279
“So it seems that “real CS” people got quite a huge result: anything that can be done in O(f(n)) compute can be done in O(sqrt(f(n))) memory. Wow. https://x.com/francoisfleuret/status/1904830459843878941
“Nvidia presents: FFN Fusion: Rethinking Sequential Computation in Large Language Models 1.71x speedup in inference latency and 35x lower per-token cost while maintaining strong performance across benchmarks https://x.com/arankomatsuzaki/status/1904370227665912243
“AMD released AMD-Hummingbird on Hugging Face Towards an Efficient Text-to-Video Model https://x.com/_akhaliq/status/1904386209373118623
“NEO at NVIDIA GTC – walking speed looks solid. [no sound] https://x.com/TheHumanoidHub/status/1902820130234753487
“🚀 NVIDIA’s Canary models are going brrrrr 🏎️💨 🔥 Canary-1B-Flash is 4.44× faster than Canary-1B with similar performance!(6.35% vs. 6.5% WER) ⚡ Canary-180M-Flash is 5.24× faster while only dropping less than 10% in WER (7.12% vs. 6.5%) https://x.com/Tu7uruu/status/1902726741774279122
Former Intel CEO Gelsinger joins religious-oriented tech firm Gloo for AI push | Reuters https://www.reuters.com/technology/former-intel-ceo-gelsinger-joins-religious-oriented-tech-firm-gloo-ai-push-2025-03-24/
Nemotron-H: A Family of Accurate, Efficient Hybrid Mamba-Transformer Models – NVIDIA ADLR https://research.nvidia.com/labs/adlr/nemotronh/
“A crucial point that everyone should be internalizing: in the age of test-time search, it’s pretty much always possible to reach any level of capability by simply expending more compute. So it’s not just about, “can you do it?”. The key is how efficiently you can do it.” / X https://x.com/fchollet/status/1904267900963475807
Introducing Instella: New State-of-the-art Fully Open 3B Language Models — ROCm Blogs https://rocm.blogs.amd.com/artificial-intelligence/introducing-instella-3B/README.html
“Nvidia just dropped FFN Fusion Rethinking Sequential Computation in Large Language Models https://x.com/_akhaliq/status/1904390303458459821
“1X and NVIDIA collaborated to train a model based on the open-source GR00T N1 that enables NEO to load a dishwasher autonomously. The 1X AI team developed key software components, including a dataset API that enabled NVIDIA to access real-world data collected by 1X and an https://x.com/TheHumanoidHub/status/1903095238505402566
“New on LMArena: @Nvidia’s Llama-3.3-Nemotron-Super-49B-v1 lands at #14! A powerful open reasoning model—top-15 overall, excelling in math, with an openly released 15M post-training dataset. Congrats to the @NvidiaAI Nemo team for this fantastic contribution to the open https://x.com/lmarena_ai/status/1903116426535375060
“NEO update tomorrow at 7:00 AM PT. What’s going on with that NVIDIA branded jacket?” / X https://x.com/TheHumanoidHub/status/1902936632925491454
“WPP, the world’s largest advertising company, is collaborating with NVIDIA, Boston Dynamics, and Canon to explore robotic-assisted cinematography and virtual ad production. Here are some cool shots of Atlas on the set. https://x.com/TheHumanoidHub/status/1903173205155815431
“Physical AI Deep Dive: Robotics Meet Reality NVIDIA’s Multi-Trillion Dollar Vision. Google & Tesla’s Moat. Building World Models & Simulating Reality. 🔗 33 min deep dive: https://x.com/bilawalsidhu/status/1903880596654567863
“Ranking overview of this model. It had been previously testing under the codename “march-chatbot” on LMArena. Check out Llama Nemotron and other frontier models at: https://x.com/lmarena_ai/status/1903116429508886904
