GLM-5.2 is free on Hugging Face Inference Providers via Zai, Together AI, Novita, Fireworks, DeepInfra for the next 6 hours. Configure it with Pi, opencode, Codex, Claude Code or any coding agent.
OPEN SOURCE
-
GLM-5.2: best open-weight model with multi-head latent attention
By
–
Just updated with the recent release of GLM-5.2. The best open-weight model today. Architecture-wise, it is built on the GLM-5 and GLM-5.1 architecture that I covered previously, meaning it reuses the Multi-head Latent Attention mechanisms.
-
Post sharing links to Deepeval and LangChain
By
–
Docs: https://
deepeval.com/integrations/f
rameworks/langchain#in-cicd-pytest
… GitHub Repo: https://
github.com/confident-ai/d
eepeval
… -
Pytest for AI Agents: Testing LangChain chains locally
By
–
Pytest for AI Agents! (100% open-source and runs locally) Building agents with LangChain means chaining LLMs, tools, and retrieval steps together. Each component can fail differently. The output changes with every run. Traditional unit tests don't work here because there's no
-
Nuance in ‘open-weight model’: OpenRouter vs Ollama
By
–
In general, I agree. But they are related but slightly different things in some contexts. Like when I say "I am running an open-weight model", I could be using OpenRouter or ollama cloud models.
-
OpenKB solves RAG’s amnesia problem based on Karpathy’s concept
By
–
TRADITIONAL RAG HAS A MASSIVE AMNESIA PROBLEM. It rediscovers knowledge from scratch on every single query, and nothing ever accumulates. OpenKB is a new open-source alternative that finally fixes this. Based on a brilliant concept outlined by @Karpathy
, OpenKB treats -
Google Gemma team open-sources cookbook with concurrent apps
By
–
repo → https://
github.com/google-gemma/c
ookbook/tree/main/apps/concurrent
… Huge shoutout to the @googlegemma team for building this and open-sourcing it for the community Don’t forget to drop a ! -
Gemma 4 26B demo: one orchestrator, 10 parallel agents, all local
By
–
🚨 One orchestrator. 10 parallel agents. 100+ tokens a second.
— Charly Wargnier (@DataChaz) 18 juin 2026
All local.
The @googlegemma team just dropped a MASSIVE demo for Gemma 4 26B.
They built a concurrent workflow that lets the 26B model coordinate an entire team of sub-agents on your machine.
Out of the box, the… pic.twitter.com/KoljxPbJheOne orchestrator. 10 parallel agents. 100+ tokens a second. All local. The @googlegemma team just dropped a MASSIVE demo for Gemma 4 26B. They built a concurrent workflow that lets the 26B model coordinate an entire team of sub-agents on your machine. Out of the box, the
-
VivaTech: France transforms AI ambitions into reality
By
–
At VivaTech, we share how France transforms AI ambitions into production reality. AI factories come into service, open models progress, and agents are deployed across various sectors, including healthcare, telecommunications, the
-
Gigatron for Linear Algebra in Machine Learning
By
–
Gigatron for Linear Algebra in Machine Learning! @gp_pulipaka
! #BigData #Analytics #DataScience #AI #MachineLearning #IoT #IIoT #PyTorch #Python #RStats #TensorFlow #Java #JavaScript #ReactJS #GoLang #CloudComputing #Serverless #DataScientist #Linux #Mathematics #Programming