2026-10-03 · 8 min
TAG
Benchmarks
20 posts tagged “benchmarks”
2026-09-18 · 10 min
Mac vs mini-PC vs DGX Spark vs DIY: the local-AI platform decision, measured where possible
2026-09-18 · 11 min
The $1.5k to $6k local AI build: measured tiers, not part lists
2026-09-18 · 12 min
Your Local LLM Isn't as Dumb as It Feels: The Six Config Traps, Measured
2026-07-25 · 11 min
Battlemage one year in: Arc B580/B60 as a local-AI card
2026-07-24 · 14 min
Best local LLMs in 2026: measured, not summarized
2026-07-23 · 9 min
Quantization in 2026: Q4_K_M is no longer the compromise
2026-07-22 · 9 min
The MoE shift: why every new local model is a Mixture-of-Experts
2026-07-19 · 12 min
Best GPUs for local AI in 2026: measured cross-vendor
2026-07-19 · 11 min
Best local LLM for coding in 2026: measured
2026-07-19 · 13 min
CUDA vs ROCm vs Vulkan vs OpenVINO: measured
2026-07-19 · 8 min
Dual 12GB vs single 24GB: we ran the same 27B on four machines and the GPU count barely mattered
2026-07-19 · 11 min
Ollama vs llama.cpp on the same hardware: measured
2026-07-19 · 11 min
Single vs dual GPU for local LLMs: when the second card stops idling
2026-07-19 · 12 min
Speculative decoding on consumer GPUs: MTP measured 2.07×
2026-07-19 · 10 min
What a quantization tier costs in watts: measured
2026-07-18 · 9 min
5 things you can do with a local LLM today: Practical setups with measured performance
2026-07-18 · 9 min
Arc B60 vs RX 7900 XT: a real comparison, not a chart fight
2026-07-18 · 7 min
OpenVINO beats Vulkan on the Arc B60 — and we were wrong about SYCL
2026-07-18 · 10 min