Fun new game: prompt gymnastics to get past Sora’s content filter 😂 Here’s one result:

E.g., see this recent anthropic status update. “Lower intelligence, malformed responses, issues with tool calling in Claude code” — quantized models is one way this could happen.

Used Claude Code to build a simple tool tmux-cli that lets CC puppet your terminal -- i.e. interact with CLIs expecting user input -- useful to let CC autonomously run debuggers, test CLI/UI scripts, spawn other Claudes (fully visible unlike black-box sub-agents) github.com/pchalasani/c...

The work is mysterious and important:

lumon-industries.com

QwQ-32B-Preview from Qwen is the best "o1-like" LLM today. You can run it remotely via glhf.chat super-fast endpoint.

Langroid lets you easily use it -- it gets the "Sally question" right, after a LOT of thinking 🤔 (images below)

Langroid tour: langroid.github.io/langroid/tut...

Some impressive stats from the groq discord:

llama3.3-70b is a seriously impressive model. Langroid lets you easily try it via groq's insanely fast endpoint - just set the model name to "groq/llama-3.3-70b-specdec"

Langroid: agent-oriented LLM framework from CMU/UW-Madison researchers
Quick tour: langroid.github.io/langroid/tut...

So I dug a bit into aisuite to see how it converts Anthropic's response to OpenAI format, and it looks much simpler than I expected (see img) -- then I realized this is because it doesn't support streaming responses.

one of the nice things about bsky is how easy it is to get data from it via the "atproto" python client. See the image and gist for example code, showing how you can get an author's posts, and get your own "home" timeline.

gist: gist.github.com/pchalasani/c...

My deepseek run gave the wrong answer :)