Ultralytics’ Post

Great guide on optimizing Ultralytics YOLOv8 models for NCNN format – essential reading for anyone looking to deploy on mobile or embedded systems. #YOLOv8 #ComputerVision #ModelOptimization

Export Ultralytics YOLOv8 to NCNN 🔥🔥🔥 Deploying computer vision models on low-power devices like mobile or embedded systems requires optimization for optimal performance. The export to NCNN format in Ultralytics YOLOv8 facilitates this, making models suitable for various lightweight applications on mobile and embedded devices. What's Included 😻 • Key Features of NCNN Models • Deployment Options with NCNN • Export to NCNN: Converting Your YOLOv8 Model Explore More ➡ https://lnkd.in/dzVnazB2 Kudos to Abirami Vina for articulating this effectively :) #computervision #objectdetection #yolov8 #researchanddevelopment #ncnn

Our new Ultralytics Tencent NCNN export Docs page by Abirami Vina. In some environments like our GitHub ARM64 macOS M1 runners NCNN appears to be the fastest format we offer. #ultralytics #yolo #ai

With a RISC-V core, two NPU coprocessors, and a DSP, this tiny chip draws under 260μW while performing on-device inference. Full article ➡ https://lnkd.in/dhyReZ-9 #artificialintelligence #EdgeAI #FullStack #machinelearning  

"AlterEgo is a non-invasive, wearable, peripheral neural interface that allows humans to converse in natural language with machines, artificial intelligence assistants, services, and other people without any voice—without opening their mouth, and without externally observable movements—simply by articulating words internally. ... A primary focus of this project is to help support communication for people with speech disorders including conditions like ALS (amyotrophic lateral sclerosis) and MS (multiple sclerosis). Beyond that, the system has the potential to seamlessly integrate humans and computers—such that computing, the Internet, and AI would weave into our daily life as a "second self" and augment our cognition and abilities." https://lnkd.in/gK-jJHnz

AI in the palm of your hand.

Check out the new blog post from Quadric about how the Chimera GPNPU can help future-proof your AI/ML SoC for the ML networks that will inevitably come down the road.

Introducing YOLOBench: Benchmarking Efficient Object Detectors on Embedded Systems

Now you can speed up your large language model (LLM) fine-tuning process using "Unsloth." Unsloth is a breakthrough library designed to work seamlessly with HuggingFace, enhancing the efficiency of LLM fine-tuning on NVIDIA GPUs. You can leverage Unsloth for fine-tuning different architectures, including "Llama" and "Mistral," using the TRL trainers (SFTTrainer, DPOTrainer, PPOTrainer). I tried to fine-tune a Mistral 7B LLM (4bit) on the IMDB dataset for text generation, all within Google Colab using Unsloth. Here's the link to my hugging face model - https://lnkd.in/gJ6Bxb9X Unsloth github - https://lnkd.in/gSxk5s96

If you work at the intersection of #AI and #infrastructure, you need to read this blog by VAST Data's very own Dr. Subramanian Kartik, Ph. D. As with any other workload, understanding true application (or in this case, #LLM model size) requirements will help ensure optimal performance while reducing spend. "The bottom line of this example is that the size of a checkpoint doesn’t depend at all on the size of the data or the number of GPUs, only on the size of the model."