Foundation Model

DeepInfer: High-performance, large-scale inference framework for DeepSeek models

Why your local LLM feels dumber than it is

Everyone is building LLM routers, we deprecated ours.

Kimi K3 Architecture Notes

Nativ — Run AI locally on your Mac

Is Grep All You Need? How Agent Harnesses Reshape Agentic Search

Introducing Gemma 4 12B

Introducing Unsloth Studio | Unsloth Documentation

Emotion Concepts and their Function in a Large Language Model

Nicholas Carlini - Black-hat Large Language Models | [Un]prompted 2026 - YouTube

Introducing Claude Opus 4.7, Anthropic

Embarrassingly Simple Self-Distillation Improves Code Generation

Show HN: 1-Bit Bonsai, the First Commercially Viable 1-Bit Large Language Models

PrismML — Concentrating Intelligence

GitHub - z-lab/paroquant: [ICLR 2026] ParoQuant: Pairwise Rotation Quantization for Efficient Reasoning in Large Language Model Inference

Conditional Memory via Scalable Lookup: A New Dimension of Sparsity for Large Language Models

NVIDIA PersonaPlex: Natural Conversational AI With Any Role and Voice - NVIDIA ADLR

Ask HN: What is the best way to provide continuous context to models?

Recursive Language Models

Reimagining LLM Memory: Using Context as Training Data Unlocks Models That Learn at Test-Time

We present Olmo 3, our next family of fully open, leading language models

A2UI

Nano Banana Pro: Gemini 3 Pro Image model from Google DeepMind

Supercharge your OCR Pipelines with Open Models

Gemini 3: Introducing the latest Gemini AI model from Google

said we should delete tokenizers

"🚀 Hello, Kimi K2 Thinking! The Open-Source Thinking Agent Model is here"

Source: Thanks and Bharat for showing the world you can in fact tra...

Tongyi DeepResearch: A New Era of Open-Source AI Researchers | Tongyi DeepResearch

MiniMax-M2

I quite like the new DeepSeek-OCR paper

olmOCR 2: Unit test rewards for document OCR | Ai2

How to Get Consistent Classification From Inconsistent LLMs? "How to Obtain Consistent Classification From Inconsistent Language Models?"

Stanford's ALL FREE Courses [2024 & 2025] ❯ CS230 - Deep Learni...

PaddleOCR-VL: Boosting Multilingual Document Parsing via a 0.9B Ultra-Compact Vision-Language Model

NeuTTS Air

My trick for getting consistent classification from LLMs

EU-funded TildeOpen LLM delivers European AI breakthrough for multilingual innovation | Shaping Europe’s digital future

Qwen-Image

Huge AI market opportunity in 2025

dots.ocr: Multilingual Document Layout Parsing in a Single Vision-Language Model

Small models are the future of agentic ai

Kimi K2: Open Agentic Intelligence

Introducing Qwen3-Max-Preview (Instruct)

VibeVoice: A Frontier Open-Source Text-to-Speech Model

Apertus 70B: Truly Open - Swiss LLM by ETH, EPFL and CSCS

Alexander Kruel - Links for 2025-08-24

DSPy

Build a Large Language Model (From Scratch)

CS294/194-196 Large Language Model Agents | CS 194/294-196 Large Language Model Agents

Claudia – Desktop companion for Claude code

The race for LLM cognitive core

Qwen3-Coder: Agentic coding in the world

Voxtral | Mistral AI

SymbolicAI: A neuro-symbolic perspective on LLMs

Gemini for Google Workspace Prompting Guide 101

MCP is eating the world—and it's here to stay

Building Effective AI Agents

Snorting the AGI with Claude Code

Nanonets-OCR-s – OCR model that transforms documents into structured markdown

[2505.24863] AlphaOne: Reasoning Models Thinking Slow and Fast at Test Time

[2505.24864] ProRL: Prolonged Reinforcement Learning Expands Reasoning Boundaries in Large Language Models

Ask HN: What is the best LLM for consumer grade hardware?

Show HN: AutoThink – Boosts local LLM performance with adaptive reasoning

Show HN: My LLM CLI tool can run tools now, from Python code or plugins

A Research Preview of Codex

Ollama's new engine for multimodal models

Vision Now Available in Llama.cpp

Token & Token Usage | DeepSeek API Docs

Gemma 3 QAT Models: Bringing state-of-the-Art AI to consumer GPUs

GitHub - HandsOnLLM/Hands-On-Large-Language-Models: Official code repository for the O'Reilly Book - 'Hands-On Large Language Models'

DeepSeek-R1 incentivizes reasoning in LLMs through reinforcement learning | Nature

A foundation model to predict and capture human cognition | Nature

Large language models are proficient in solving and creating emotional intelligence tests | Communications Psychology