Btw the model router is the amazing open model Arch-Router-1.5B by @katanemo_ @salman_paracha
: https://
huggingface.co/katanemo/Arch-
Router-1.5B
…
LLMS
-

Arch-Router-1.5B: Open Source Model Router by Katanemo
By
–
-
LangGraph: Building Production-Ready AI Agents with Minimal Abstraction
By
–
In this blog piece, you’ll learn why and how we built LangGraph for production agents. Building upon feedback from the super popular LangChain framework, we aimed to find the right abstraction for AI agents, and decided that was little to no abstraction at all. Instead, we
-

Transformers Learning Multiplication Through Auxiliary Loss
By
–
Why Can't Transformers Learn Multiplication? This paper found that plain training never builds long-range links of multiplications. So by adding a new auxiliary loss that predicts the “running sum”, it enables the model to successfully learn multi-digit multiplication!
-
Claude Multi-Agent Systems: Skills, Code, and Future Insights
By
–
Very interesting convo with @ErikSchluntz (multi-agent research lead) about all things agents:
— Alex Albert (@alexalbert__) 17 octobre 2025
– Why Claude is good at agent tasks
– Evolution from workflows to multi-agents
– Tips for creating Skills
– How code unlocks all other domains
– How to use subagents
– The future of… pic.twitter.com/NshRStZVdsVery interesting convo with @ErikSchluntz (multi-agent research lead) about all things agents:
– Why Claude is good at agent tasks
– Evolution from workflows to multi-agents
– Tips for creating Skills
– How code unlocks all other domains
– How to use subagents
– The future of -

Stanford Paper Makes Prompt Engineering Obsolete Forever
By
–
Stanford just killed Prompt Engineering — and that’s a good thing.
A single paper just changed how we understand alignment, diversity, and creativity in AI forever. — RIP Prompt Engineering A groundbreaking new Stanford paper just made it obsolete — with one simple -

GPT-5 Breakthrough: Multi-Model Message Routing for Optimization
By
–
The main breakthrough of GPT-5 was to route your messages between a couple of different models to give you the best, cheapest & fastest answer possible. This is cool but imagine if you could do this not only for a couple of models but hundreds of them, big and small, fast and
-
PaddleOCR VL 900M Params Achieves SOTA on OmniDocBench
By
–
Pretty wild what 900M params can do – PaddleOCR VL > SOTA on OmniDocBench v1.0 & v1.5 (text, tables, formulas, charts, reading order)
> Multilingual – 109 languages (Latin, Arabic, Cyrillic, Devanagari, Thai, etc)
> Handles handwriting, historical docs, noisy scans
> Supports -

Bee: Open-Source Multimodal LLM Corpus and Framework
By
–
Bee A High-Quality Corpus and Full-Stack Suite to Unlock Advanced Fully Open MLLMs
-
Claude’s Dynamic and Composable Expertise
By
–
1. Dynamic Expertise
— God of Prompt (@godofprompt) 17 octobre 2025
Claude now scans your available skills and automatically loads only what’s relevant to the task.
No bloat. No latency. Just the right expertise, when needed.
2. Composable + Portable
Skills stack together like Lego pieces.
Build once → use across Claude… pic.twitter.com/EkFYL92IeN1. Dynamic Expertise Claude now scans your available skills and automatically loads only what’s relevant to the task. No bloat. No latency. Just the right expertise, when needed. 2. Composable + Portable Skills stack together like Lego pieces.
Build once → use across Claude
