People on Social Media are excited about a new model on the LMSYS Chatbot Arena: gpt2-chatbot. Some theorize that it may be GPT-2. I have fingerprinted the Tokenizer, and no: not GPT-2, but consistent with OpenAI cl100k (used for GPT 3.5 onwards). Peculiar gaps in world knowledge (both niche and common knowledge) were the same as in the other GP... Read more 29 May 2024 - less than 1 minute read
Inspired by some recent discussion about website optimization, I have posed a query for factual background information to Bing Copilot - which I have recently come to like. In my tests, this pulled background information from a relevant Confluence and Service Desk, but mixed up two products by the same company - while I was actually looking for ... Read more 26 May 2024 - 1 minute read
description: “Introduces FluxText reactive text editing framework, detailing collaborative APIs, performance optimizations, and real-time synchronization features.” layout: post title: “FLUX.1: Examining Text Rendering Capabilities in AI Image Generation” date: 2024-08-25 last_updated: 2024-08-25 tags: [ai, text-to-image, flux1, black-forest-lab... Read more (Updated) - 2 minute read
Ethan Mollick recently shared some intriguing data about AI spending patterns. The numbers, sourced from a Ramp report, seem to paint a rosy picture for OpenAI. But it’s worth digging a bit deeper. According to the report, OpenAI is experiencing impressive retention and growth numbers. A whopping 82% of companies that spent on OpenAI a year ago... Read more 17 May 2024 - 1 minute read
OpenAI have updated the API cookbook to walk through the basics of multimodality and using GPT-4o via the API. However, the code sample there proves that GPT-4o does not (yet?) process video natively and instead relies on images extracted. The code sample does this at a fixed sampling rate of 0.5 Hz (so every two seconds). My question if there a... Read more 15 May 2024 - less than 1 minute read