PinnedOutperforming Claude 3.5 Sonnet with Phi-3-mini-4k for graph entity relationship extraction tasksWhen you need fast and high-throughput graph extraction with better quality than Claude 3.5 Sonnet.Aug 154Aug 154
PinnedContext is King —Evaluating real-time LLM context quality with RagasWe explore the Answer Correctness and Context Precision for the top context providers in the LLM space, JinaAI, Tavily, and AskNews.Jun 102Jun 102
The many faces of Fake NewsIf you believe the critics on the Internet, the world is plagued by a new problem called Fake News that makes trusting news difficult.Jun 28Jun 28
AskNews’ Transparency ProtocolsCombating bias in news is no easy feat. Here is how we are tackling this challenge.Jun 281Jun 281
Infusing any LLM with news — one line of codeIf you your LLM needs fast access to up-to-the minute news context, AskNews has a single line of code for you 🚀May 2May 2
Ray vs Dask: Lessons learned serving 240k models per day in real-timeReal-time, large-scale model serving is becoming the standard approach for key business operations. Some of these applications include…Aug 22, 20231Aug 22, 20231
Real-time head-to-head: Adaptive modeling of financial market data using XGBoost and CatBoostWhich gradient boosted decision tree algorithm is superior for adaptive modeling of financial market data, XGBoost or CatBoost? There are…Jun 16, 20233Jun 16, 20233
FreqAI — from price to predictionA guide to feature engineering for algorithmic trading using machine learning - by Robert Caulk and Elin TörnquistMay 16, 20232May 16, 20232