Viewing profile — bcarambio

Open Source LLM conversational memory using OctoAI and Pinecone, with code examples. If you saw yesterday's announcement from OpenAI about their limited access memory feature for C…

I started a project to snapshot the 10 most popular LLMs from the last six months. As expected, lots of churn, big hype cycles for Llama2 and Mixtral. Mixtral looks like it might s…

octoml-profile is a python library and cloud service designed to provide the simplest experience for assessing and optimizing the performance of PyTorch models on cloud hardware. W…

Hi from OctoML -- our team just put out some benchmarking results using GPT-J with a few model config/hw combinations that have relatively low latency: https:/octoml.ai/blog/evalua…

Hi from OctoML -- our team just put out some benchmarking results using GPT-J with a few model config/hw combinations that have relatively low latency: /octoml.ai/blog/evaluating-g…