Category: Local LLM
How to Run Gemma 4 Locally (2026): Which Size Fits Your GPU, From E2B to 31B
Run Google’s Gemma 4 on your own PC with Ollama, llama.cpp or LM Studio. Exact download sizes for E2B, E4B, 12B, 26B A4B and 31B, the new QAT…
Mac vs PC for Local AI in 2026: Mac Studio M5 Max and M5 Ultra vs RTX GPUs vs DGX Spark vs Ryzen AI Max
Mac vs PC for local AI in 2026: Mac Studio M5 Max and M5 Ultra unified memory vs RTX 5090 and 32 GB GPUs vs DGX Spark and…
Best Workstation GPU for AI in 2026: RTX PRO 6000 vs RTX PRO 5000 72 GB vs Radeon AI PRO R9700 vs Arc Pro B70
The best workstation GPUs for AI in 2026 compared: RTX PRO 6000 Blackwell 96 GB, RTX PRO 5000 48/72 GB, RTX PRO 4500 and 4000, Radeon AI PRO…
How Much VRAM Do I Need for AI? LLM, Image and Video VRAM Requirements (2026)
How much VRAM you need for local AI: the formula for LLM weights and KV cache, a 4B-120B VRAM table, MoE offload, and real requirements for FLUX 3…
n8n + Ollama: Build a Free Local AI Agent With Self-Hosted n8n (2026 Step-by-Step Guide)
How to connect n8n to Ollama and build a free, private AI agent on your own PC in 2026: Docker setup, the Ollama credential URL that actually works,…
Best GPU for Local LLMs in November 2026: Picks by VRAM and Real Street Prices
The best GPU for local LLMs in November 2026, ranked by VRAM with real street prices: RTX 5060 Ti 16 GB, R9700 and Arc Pro B70 32 GB,…
Tencent Hy4 Launched: Vision, Computer Use and Open Weights Coming (Benchmarks and Pricing)
Tencent’s Hy4 adds vision and computer use (trained on Blender and CAD), cuts hallucinations to 7.2% and costs about $0.90/M input tokens. Open weights are promised in days.…
How to Delete Ollama Models and Fully Uninstall Ollama (Windows, Mac, Linux, PowerShell)
Ollama models are big. One 27B model is 18 GB, a coding MoE is 23 GB, and after a few weeks of testing new releases it is easy…
Run Xiaomi MiMo-V2.6 9B Locally: Ollama, llama.cpp and LM Studio Setup (8 GB GPU)
Xiaomi’s MiMo-V2.6 9B (full name MiMo-V2.6-Distill-Qwen-9B) landed in the official Ollama library yesterday, and it is the most interesting small model of the autumn. Xiaomi took Qwen3.5 9B…
Best Ollama Models in 2026 by VRAM: Coding, Chat, Vision and Agents (Ranked)
The best Ollama models in November 2026 are not the ones most “best of” lists still recommend. Llama 3, Mistral 7B and Phi-3 are years old now. In…