Production Guides
Lately, I am feeling more and more confident about Portkey solving *real* production challenges for our users. Of course, developer love and revenue increase are important parameters, but there's something else just as high signal: The platform modularity that customers demand from Portkey, and how fast we serve
Production Guides
2024 is the year where Gen AI innovation and productisation happens hand-in-hand. We are seeing companies and enterprises move their Gen AI prototypes to production at a breathtaking pace. At the same time, an exciting new shift is also taking place in how you can interact with LLMs: completely new
AI
Retrieval-Augmented Generation (RAG) models represent a fascinating marriage of two distinct but complementary components: retrieval systems and generative models. By seamlessly integrating the retrieval of relevant information with the generation of contextually appropriate responses, RAG models achieve a level of sophistication that sets them apart in the realm of artificial
Production Guides
In two days (i.e. Jan 4), OpenAI will retire 33 models, including GPT-3 (text-davinci-003) and various others. This is OpenAI's biggest model deprecation so far. Here's what you need to know: GPT-3 Model Retirement The text-davinci-003 model (commonly known as GPT-3) will be unavailable from
Production Guides
The landscape of AI development is rapidly evolving, and open-source Large Language Models (LLMs) have emerged as a key foundation for building AI applications. Anyscale has been a game-changer here with their fast and cheap APIs for Llama2, Mistral, and more OSS models. But to harness the full potential of
Production Guides
In this blog post, we explore a roadmap for building reliable large language model applications. Let’s get started!
Production Guides
Implementing semantic cache from scratch for production use cases.
Production Guides
Learn how to use the eval framework to evaluate models & prompts to optimise LLM systems for the best outputs.
Production Guides
Choosing an LLM from 50+ models available today is hard. We explore Elo ratings as a method to objectively rank and pick the best performers for our use case.