We have successfully fine-tuned a Llama-3-8B-Instruct model on a custom dataset for function calling to the CoinGecko API. 

One of our lead developers, Andrii Kotliar, stated “Creating the dataset took some time as we figured out how to ensure the model responds appropriately

Today we're excited to release Sanctum v1.6 with a new UI and a set of new features for the dev community ✨ 🎭 New UI With simplified navigation and updated interface Sanctum gets better, making room for even more exciting features. 🗄️ Local API Server Powered by llama.cpp,

Cooking something really COOL for the next release 👨‍🍳

Excited to share that Sanctum is sponsoring the AI Privacy & Safety night at Massachusetts Institute of Technology (MIT) on Nov 20th w/ The AI Furnace 🧨🔥! I'll be demo-ing our app alongside other awesome companies, including Enkrypt AI, Versara AI & I-XRAY. Come by if you're in Boston: lu.ma/yj4d7vfs

We're excited to announce that Sanctum is joining the QUBIC Labs family with a $25K grant through our R&D Innovation Program!🚀🤝 Sanctum is revolutionizing #AI privacy with: • Local open-source LLMs • On-device encryption • Offline AI capabilities Join us in

Our application is LIVE on Berachain Foundation 🐻⛓ 🪇🎀🪇

Lending onchain is very inefficient. Here's why: - Most lent capital is unused and could more effectively be deployed elsewhere - Limited price discovery between lenders/borrowers - Risk management is at odds with liquidity Dahlia solves it. Built on Royco. Pretty exciting