Best Used GPUs for Local AI in 2026 (September 2026)
The RTX 30 and 40 series cards that still clear current Ollama and llama.cpp floors, what VRAM tier each opens, and which variants are worth the used premium.
Tag
#hardware
What to Check Before Buying a Used GPU for Local AI
The compute capability, driver and ROCm facts to verify on a second-hand card before it has to run Ollama, llama.cpp or a current CUDA toolkit.
Best GPUs for Local AI (August 2026)
VRAM decides what runs at all, bandwidth decides how fast it types. A buying guide by tier across NVIDIA, AMD, Intel, Apple and the used market.
Kimi K3 Weights Are Public: 2.8T Params, 64-GPU Reality
Moonshot released the full Kimi K3 weights on July 27, 2026: 2.8T params, 1M context, MXFP4. Read the license before you plan a deployment.
Apple's Trade-Secret Suit Against OpenAI, Explained
Apple's July 10 lawsuit names Tang Tan, Chang Liu, and 400-plus ex-Apple hires. Here's what the complaint alleges and what it hits at OpenAI.
Arm Releases Its First Chip in 35 Years: A 136-Core Data Center CPU Co-Built With Meta
Arm shifts from licensing IP to selling silicon, betting its AGI CPU can deliver 2x performance per rack versus x86 and save hyperscalers billions.
The Great AI Bifurcation: Hardware Giants Soar While Enterprise Software Stumbles
Enterprise AI spending is up 35%, but the money is flowing to chips and infrastructure while software stocks crater
THOR Brings Brain-Inspired Computing to Any Researcher
The first US open-access neuromorphic computing hub launches with 400,000 processors simulating 393 million neurons
Nvidia GTC 2026 Preview: Everything Expected From Jensen Huang's Monday Keynote
From the Vera Rubin architecture to NemoClaw enterprise agents, here's everything Nvidia is expected to unveil at its biggest conference of the year
Taalas HC1: The AI Chip That Bakes the Model Into Silicon
A Toronto startup is etching LLM weights directly into transistors, achieving 17,000 tokens per second. The catch: you can't change the model.
Local AI Showdown: Best Open-Weight Models for Your Hardware
A tier-by-tier comparison of the top open-weight LLMs you can run locally, from 8GB laptops to 24GB gaming GPUs to Apple Silicon Macs.
OpenAI Just Deployed Its First AI Model on Non-NVIDIA Chips
GPT-5.3-Codex-Spark runs on Cerebras' wafer-scale chips at 1,000+ tokens per second. It's OpenAI's first production break from NVIDIA - and it won't be the last.
Tiiny AI Pocket Lab: Scrutinizing Its 120B Claim
Tiiny AI says its 300-gram Pocket Lab runs 120B models locally. The design is plausible, but performance and privacy claims remain unverified.