# VRAM usage

The AI Whale Fall: How Big Players' Training Costs Nourish the Open Source Ecosystem

The "AI Whale Fall" reveals a hidden industry symbiosis: giants like OpenAI and Meta spend billions training models, and their open weights and research continuously nourish the open source ecosystem.

Running Qwen 122B on 8GB VRAM: A Complete Guide to Quantized Local Deployment

A Bilibili creator ran Qwen 122B with 256K context on just 8GB VRAM + 64GB RAM using llama.cpp. Full breakdown of quantization, deployment params, performance, and cost-effective alternatives.

LoRA Training Showdown: Which Base Model Is Best for Character Fine-Tuning? A 6-Model Real-World Test

Real-world LoRA training comparison across Ideogram, Flux 1 Dev, Z Image, Flux 2 Klein, and Krea — revealing which base model best handles face fidelity and generalization for AI portrait developers.

The Complete Optimization Guide for Local AI Image & Video Generation with 16GB VRAM

What can 16GB VRAM do? This guide covers FLUX, SDXL, Wan video models, ComfyUI workflows, GGUF quantization, and VRAM optimization to max out your RTX 16GB GPU.

Jul 17, 2026·3 min

AI Agent Home Server Selection Guide: Low-Cost OpenClaw Deployment in Practice

Discover why AI Agents burn through API budgets fast, and how to deploy OpenClaw on a home server using Ollama, DeepSeek, and Gemini in a cost-effective hybrid setup.

ComfyUI v0.28.0: Native Video Super-Resolution, 3D Support, and PixelDiT Architecture — A Complete Breakdown

ComfyUI v0.28.0 adds SeedVR2 native video super-resolution, PixelDiT architecture, 3D Gaussian Splatting export, int4 quantization, and lip-sync integration for a major multimodal AI workflow upgrade.

Inkling Open-Weights Model: A New Expl…

Inkling Open-Weights Model: A New Exploration in the Open-Source AI Ecosystem

Inkling open-weights model released. Explore the key difference between open-weights and fully open-source AI, Inkling's positioning, and how to choose the right open-source LLM for production.

Bonsai-27B Extreme Quantization: Running 1-bit LLMs on Consumer Hardware

Bonsai-27B supports binary/ternary extreme quantization for 27B LLMs running on 8GB VRAM. Covers llama.cpp upstream progress, RTX 4060 benchmarks (30 t/s), and real-world limitations.

GPUStack in Action: Complete Guide to Day 0 Deployment of DeepSeek-V4-Flash

Deploy DeepSeek-V4-Flash DSpark on 8× H20-141G using GPUStack's SGLang backend on Day 0. Full walkthrough of Web UI config, parameter tuning, and 200 tokens/sec benchmark results.

Qwen3.5 Uncensored Version Local Deployment: Run a 35B Model on Just 6GB VRAM

Run Qwen3 35B-A3B uncensored locally on just 6GB VRAM using llama.cpp and GGUF quantization. Full deployment guide, code generation tests, and Agent integration walkthrough.

LongCat 2.0: Meituan Trains a 1.6 Trillion Parameter LLM on Domestic Chips

Meituan open-sources LongCat 2.0, a 1.6T-parameter MoE model trained on 50,000+ custom chips without NVIDIA GPUs or Google TPUs, rivaling OpenAI and Google.

ExLlamaV3 v1.0.0 Official Release: Faster Inference, Lossless KV Quantization, and Streamlined Dependencies

ExLlamaV3 v1.0.0 releases with lossless KV cache quantization via kernel fusion, removal of flash-attention-2/xformers, major GEMM/GEMV gains, and broader tensor parallelism support.

How Large a Local LLM Can You Run on a Laptop with 8GB VRAM? A Configuration Guide

Complete guide to running local LLMs on a laptop with 8GB VRAM: real usable memory, quantization estimation, Q4 7B/8B model recommendations, Ollama setup, GPU offloading, and agent development tips.

GPU Sharing in Kubernetes Production: A Deep Dive into MIG, MPS, and Time-Slicing

A deep dive into MIG, MPS, and Time-Slicing GPU sharing solutions for Kubernetes production. Compare isolation, performance, and use cases to make the right choice.

Complete Practical Guide to LLM Fine-Tuning: A Roadmap for Beginners

A practical LLM fine-tuning roadmap for beginners — covering when to fine-tune, LoRA/QLoRA selection, data prep, tools like Unsloth, and evaluation for Llama, Mistral, and Gemma.