Best Value GPU for LLM Inference (2026): Price per GB and Tokens per Dollar
The best value GPU for LLM inference depends on how much VRAM you need: the RTX 5060 Ti gives the most speed per dollar, a used RTX 3090 is the…
BUFFER’S NOTEBOOK
Category: GPUs
6 articles
AI, hardware and tools I’ve tried for myself. When a conversation on X needs more room, it continues here.
Follow via RSS
What Models Can RTX 4090 Run? An Honest 2026 Guide to 24GB VRAM
An RTX 4090 runs 8B and 14B models with room to spare, 32B at Q4 up to about 16K context, and no 70B on a single card. The VRAM math,…
RAM Price Trend 2026: Dated DDR4, DDR5 and SSD Prices Plus the Memory Price Forecast
A living RAM and SSD price tracker: dated DDR4 and DDR5 spot, contract and retail prices, a 12-month per-GB chart, what is driving prices, forecasts from TrendForce and Citi, and…
Rubin vs MI455X Specs: NVIDIA’s Quiet Memory Bandwidth Cut, Side by Side (2026)
Rubin vs MI455X specs from NVIDIA's and AMD's own pages. NVIDIA's Rubin page quietly went from 22 to 19.2 TB/s per GPU in early September. Full tables, sources, timeline and…
GPU Rental Prices Keep Rising, Yet Neocloud Stocks Fell 30% to 47%: The Surprising Debt Problem
GPU rental prices keep rising, but IREN, CoreWeave and Nebius fell 30% to 47% in a month. The 1-month drop vs the drop from the high, the CoreWeave bond and…
RTX 5090 for AI: Is It Worth It? The Simple, Honest 32GB VRAM Math (2026)
Is the RTX 5090 for AI worth it? The 32GB VRAM math by model and quantization, checked against a real benchmark, plus Mac Studio and DGX Spark specs.