Anyway, if you haven't explored JAX yet, you really should. It makes you completely change your expectations for ML frameworks 🙂
TOOLS
-
InVideo Partnership: Long-term Collaboration in Creative AI
By
–
Yes. I've made 2 videos about @invideoOfficial already. They're actually one of my long-term partners. I work closely with them. 🙂
-
H2O.ai and NVIDIA Enable LLM Deployment for Financial Services
By
–
.
@h2oai and NVIDIA are working together to provide an end-to-end workflow for financial institutions, using NVIDIA AI Enterprise. Companies can develop and deploy their own #LLMs to power #generativeAI use cases in #financialservices. Read the blog here. -
ChatGPT-Powered Advisor for Digital Strategy Innovation
By
–
Embrace the future of digital strategy with our new ChatGPT-powered advisor! Unlock tailored insights and innovative solutions to propel your business into the digital age. Stay ahead with @ingliguori https://
bit.ly/41euZNa
#Kenovy #DigitalStrategy #Innovation -
Extra Midjourney Tips: Keywords, Styles, Quality, Aspect Ratios, Chaos, Exclusions, URLs, Versions
By
–
3. Extra Midjourney tips: • Specify with keywords • Add relevant styles • Use –quality parameter • Experiment with aspect ratios with –ar
• Leverage –chaos for creativity • Employ –no for exclusions • Integrate photo URLs • Test different –style versions -
Create a logo in minutes with Midjourney AI
By
–
Imagine creating a logo in less time than making coffee. Midjourney AI turns your ideas into a logo reality. Hundreds of designs at your fingertips, in just minutes. Follow the thread:
-
Keras 3 with JAX Significantly Faster Than PyTorch
By
–
It is not uncommon for Keras 3 + JAX to be 3-5x faster than PyTorch models pulled from research repos on GitHub or from Hugging Face. Though the typical figure is more like 30% faster.
-
KerasCV Segment Anything: Zero-shot Image Segmentation with Prompt Points
By
–
KerasCV now includes a Segment Anything implementation, which enables you to do image segmentation using only "prompt points". No training data needed. With the JAX backend, it runs ~5x faster on GPU than the original PyTorch implementation. Told you JAX was fast. Guide:
-
GPT Store: opportunities with custom GPTs
By
–
GPT Store dropping next week. If you build GPTs, you may be able to really capitalize on writing a few custom instructions for everyday users. Will be interesting to see how this plays out.
-
Copilot Eliminates Need to Learn JavaScript and HTML
By
–
Copilot has fulfilled my dream of literally never having to learn javascript or html for the rest of my life and I am thrilled with the results.