the ferry had to stop… to let the orcas cross the channel
Peli de Halleux
@pelikhan.bsky.social
Working on genaiscript / LLM / GenAI . Formerly DeviceScript, JACDAC, MakeCode for micro:bit and others. Opinions are my own.
getting back into the video recording grove for genaiscript. working on the head-squeezing-surprised-looks youtu.be/Wya3MQRIbmE?...
Get started with GenAIScript With GitHub Codespaces & Models
YouTube video by Peli de Halleux
youtu.be
Went down the ffmpeg rabbit hole to support videos in #GenAIScript . microsoft.github.io/genaiscript/...
Let there be videos!
Add videos to your LLMs calls.
microsoft.github.io
Ported my contour shader from webgl to OBS. github.com/pelikhan/obs... . It's the thingy that adds a contour line after apply a green screen filter.
GitHub - pelikhan/obs-contour-shader: Shader that adds a contour to non-alpha screen section
Shader that adds a contour to non-alpha screen section - pelikhan/obs-contour-shader
github.com
Had fun adding support for video processing in #GenAIScript microsoft.github.io/genaiscript/...
Video Alt Text
Learn how to generate alt text for videos
microsoft.github.io
check out createai.microbit.org a super cool way to leverage the sensors on the mighty micro:bit and build custom AI recognizers! Same hardware keeps on giving!
micro:bit CreateAI
Create AI on your BBC micro:bit using movement and machine learning
createai.microbit.org
This is a neat new variant on RAG - no vectors, not even full-text search, instead showing the model a header hierarchy and giving it a tool to read the relevant sections My notes here: simonwillison.net/2024/Dec/6/r...
Roaming RAG – make the model find the answers
Neat new RAG technique (with a snappy name) from John Berryman: > The big idea of Roaming RAG is to craft a simple LLM application so that the LLM assistant …
simonwillison.net
John Berryman@jnbrymn.bsky.social · 2y agoTired of wrestling with complex RAG pipelines? 🚀 Enter Roaming RAG: a simpler way to make your LLMs find answers in well-structured docs. No vector databases, no headaches—just rich, structured context. 👉 Read how it works: arcturus-labs.com/blog/2024/11...
🤗 I've added support for Hugging Face Transformers.js in #GenAIScript. It's very convenient to have access to local LLMs with installing any extra toolchain. Nice SDK from the HF. microsoft.github.io/genaiscript/...
Hugging Face Transformers.js
Learn how to run LLMs locally using Hugging Face Transformers.js in GenAIScript, enabling you to leverage powerful language models in your JavaScript projects.
microsoft.github.io
I missed the announce that LMstudio added a headless server compatible with OpenAi. Nicez.