ai-engineering-at/llama-cpp-turboquant-guide

docs: overhaul README — architecture diagram, fixed anchors, VRAM visualization, bilingual

results: RTX 3090 consolidated — 4 runs, 15 measurements, avg -7.5% TPS

results: consolidate 4070 Laptop — 2 independent sessions, avg -4.6% TPS

results: add verified RTX 4070 Laptop benchmark + cross-GPU comparison table

results: add RTX 4070 Laptop 8GB benchmark (Llama-3.1 8B)

fix: add GitHub repo link prominently in HF Space README

Add HuggingFace Space frontmatter

results: add v2 verification run — confirms v1 within measurement variance

Initial release: TurboQuant practical guide for consumer hardware