Nils Durner's Blog Ahas, Breadcrumbs, Coding Epiphanies

[UPDATED] OBS Stream Recording

How does one record a live stream in absence, perhaps using AWS EC2? A discussion on Reddit lead me to base everything on a g5.xlarge instance. Problems: ⚒️ On the Windows Server, there initially was no soundcard. Fixed by simply installing Virtual Audio Cable (VAC)? Fixed by running Microsoft Teams in an OBS Source of type “Browser”: this will... Read more

[UPDATED] Chatbot Arena Trickery

description: “Investigates evaluation ‘trickery’ in chatbot tournaments, showing how adversarial inputs can skew leaderboard results and recommending robust testing protocols.” layout: post title: “LMSYS Chatbot Arena, a popularity contest” date: 2024-04-17 last_updated: 2024-04-17 tags: [llm, lmsys, gemini, reka] — Peter Gostev on LinkedIn rem... Read more

GenAI 'Workbenches' announced

I have announced the trinity of my chat interfaces on LinkedIn, peppered with some background reading/watching material. This is kind-of in response to a Singapore taxi driver asking “How to get into AI” and me frantically filling his OneNote. How I wish there was a single, actually good starting point! (This still isn’t.) Note about the Amazon... Read more

LLM on smartphone

Following up on our conversation on quantized models on smartphones, Stefano Fiorucci wrote a post about how to run a small language model on a smartphone: . This involves either Layla Lite App or Temux. One commenter recommended LLM Farm on iPhone. Read more

Gemini Large context & Video

RAGfluencers are of course discontent with Gemini’s very large 1 million token context window, noting the high costs associated with a large number of input tokens. “It feels like a very niche use case”. The niche for the the 1M tokens would be multi-modality in general and video in particular. My modest experiments suggest that the model does ... Read more