Microsoft’s NEW AI Plays You — UserLM-8b Is Wild

WanGP - AI Video Generation for GPU Poor - Use Locally with Any Model

Google's New Computer-Use Model: Power AI Agents for UI Interaction

LING 1T is Here - Liberal, Fast, Efficient and Open-Minded AI Model

LFM2-8B-A1B: A Mixture of Expert for Edge Intelligence: Run Locally

Use OpenAI AgentKit AgentBuilder with Ollama: Tutorial with MCP

TFree-HAT 7B: Tokenizer-Free Model: Install, Test and Understand Locally

Using SQLite as AI Vector Store for RAG: Hands-on Tutorial for Any Device

NeuTTS-Air: Voice AI on CPU: Run Locally

Apriel-1.5-15b-Thinker - Mid training is all you need! Run and Test Locally

Run Qwen3-VL 30B on CPU Locally for Images and Videos - No GPU Needed

CWM: Code Generation with World Models: New Agentic Coding: Test and Run Locally

Qwen3-VL-30B-A3B-Thinking: Best AI For Long Video Analysis: Run Locally

Qianfan-VL: Domain-Enhanced Universal Vision-Language Models: Run Locally

Qwen3-VL — The Most Powerful Vision-Language Model: Install and Test

Holo1.5 - Computer Use Agents - Interact with Real Applications - Run Locally

ModernVBERT: Missing Link for Vision RAG Pipeline: Run Locally

Chinese Chips Are Rising: TileLang + TVM vs. CUDA

Use Use Claude-Code with Ollama with Router Locally

Megrez2 - A Device Native LLM with MoE and VRAM Optimization

Granite 4 Micro: Hyper-Efficient, High-Performance Hybrid AI for Enterprise

ByteBot - AI Desktop Agent That Uses Computer Like Humans - Install Locally

LFM2 Audio: End-to-End Audio Foundation Model: Install and Test

KaniTTS: Fast and Expressive Speech Generation Model for English, Arabic, German

Claude Sonnet 4.5 Pushes Frontier: No-Fluff Video with Thorough Testing

POINTS-Reader from TenCent: Accurate OCR of Images: Run Locally

GLM-4.6 Just Dropped… Is This The New AI Beast?

LlamaCTL - Unified Serving and Routing for Llama.cpp, MLX and vLLM

NVIDIA Drops LongLive - Real-time Interactive Long Video Generation - Run Locally

KAT-Dev-32B, KAT-Coder: Code Intelligence through Scalable Agents

STOP THE PRESS: DeepSeek V3.2 is Here—New Attention Changes Everything

The Best AI Interview Question Ever (And the Only Right Answer)

oLLM - Run 80GB Model on 8GB VRAM Locally - Hands-on Demo

From Install to Insights: Klear-46B MoE in the Real World from Kwai

Moondream 3: No-Compromise Vision Model: Frontier-level Reasoning at Blazing Speed

Microsoft VibeVoice with ComfyUI for Single and Multi-Speaker Voice Cloning

Tencent Hunyuan 3D - Realism, Textures, and 3D World Models at Best

K2 Think - AI's Global March in Advanced Reasoning - Local Installation and Testing

Wan2.2-Animate-14B: Animate Characters in Videos with Expressions

Sentient ROMA: Meta-Agent Framework for High-Performance Multi-Agent Systems

MinerU 2.5 with vLLM: Extract Data from Any PDF - Easy Tutorial

Grok 4 Fast — Stress-Testing It on Elon-Scale Scheduling

PocketFlow - Build Modular Workflows and AI Agents: Hands-on Tutorial

China Now Has a CUDA-Compatible GPU: Fenghua No. 3

This AI Reads Your Screen and Clicks - ScaleCUA 32B - Run Locally

I Made Ollama Say the Wrong Thing—Qwen3Guard Caught It LIVE (LLMOps Crash Test)

Qwen3-Omni-30B Captioner — Best of All Time — Unbeatable Audio Understanding

I Installed Qwen3-Omni “Thinking”—It Does TEXT • IMAGE • AUDIO • VIDEO Together

Qwen3-Omni is Here with Image, Audio, Video, Text: Local Demo with Testing

DeepSeek V3.1 Terminus: Is This the Endgame Update? (Insane Real-World Tests)

LEANN with Ollama: RAG Everything on Your Laptop: Easy Tutorial

Jina Code Embeddings: Small but Performant Code Embedding Model

Lucy Edit Dev: Edit Your Videos with AI for Free and Locally

Goddess on a Trampoline Controls My Destiny… And I Let Her - Silly AI Model

ObjectClear: Complete Object Removal with Effects from Images - Install Locally

Billions Spent on AI… But Where’s the ROI?

TinyLettuce: Efficient Hallucination Detection with 17M Model - Run Locally

Expert Parallelism Explained! How 120B+ AI Models Run on GPUs in Seconds

Xiaomi Drops MiMo-Audio-7B: A Streaming Speech Conversation Model

🔥Create Suno-Level AI Music LOCALLY on Just 6GB VRAM! (SongBloom DPO + ComfyUI Setup)