ViewTube

🦁 NVIDIA & MIT AI research ➡️ usable apps 💚
✅ 48% SAM acceleration 📈   ✅ no quality degradation   ✅ app & UI code   ✅  #TensorRT 
Published by our distinguished engineer and MIT Professor Song Han.
📗➡️ https://console.brev.dev/notebooks/efficientvit-segmentation 

Choose from 90+ free  #GTC24   #training  technical training videos available on demand. Featuring a variety of topics on  #generativeAI , graphics & simulation,  #acceleratedcomputing ,  #datascience , and more. 🔍  https://nvda.ws/4a4eYff 

🦙🦙🦙 Optimize your Meta Llama 3  #inference  with TensorRT-LLM …and how! ✨ in only 4 steps.

📗➡️ https://console.brev.dev/notebook/llama3-tensorrtllm-deployment 

🦙 We’re excited to announce support for the Meta Llama 3 family of models in NVIDIA TensorRT-LLM, accelerating and optimizing your LLM inference performance. 
Get started ➡️  https://nvda.ws/3UtisDB 

Learn about emerging apps, societal implications, and strategies for scaling generative AI tools effectively.  Watch the discussion with Open AI COO, Brad Lightcap, and our VP Manuvir Das, in this informative  #GTC24  session. ➡️  https://www.youtube.com/watch?v=M2s17... 

😻 😸 Kitten-to-text: The visual language model ViLA is supported in the latest release of TensorRT-LLM v0.9. 👉  https://github.com/NVIDIA/TensorRT-LLM/tree/main?tab=readme-ov-file#versions-090  Our Research and Distinguished Engineer and MIT professor, Song Han, developed and incorporated ViLA into  #TensorRT   #LLM . 

🏗️ 📐 Tired of marching cubes producing skinny triangles and losing sharp geometric details?
Check out the newest NVIDIA Kaolin tutorial to learn how to add Flexicubes for mesh optimization. 
👀   https://www.youtube.com/watch?v=2gqds...   #NVIDIAResearch 

There are 2 ways to fine-tune the new Gemma models:
✅ LoRA fine-tuning, and ✅ Supervised fine-tuning
✨ Run CodeGemma on TensorRT-LLM ➡️ https://nvda.ws/3U8i4dz 


 #LLM   #GoogleCloudNext   @GoogleDevelopers 

NVIDIA and Google DeepMind have collaborated to optimize Gemma models for NVIDIA AI platforms. Experience the power of RecurrentGemma and CodeGemma for domain-specific use cases on the NVIDIA API catalog. ✨ https://t.co/v7zPyqvosX 

#CUDA : New Features and Beyond is now available on our YouTube channel. This is one of our most popular technical sessions at  #GTC24 . During this engineering-focused talk, NVIDIA CUDA Architect Stephen Jones shares what’s new and what's coming next, for both CUDA and GPU computing‌. 
➡️  https://youtu.be/pC0SIzZGFSc