本文へ

Blog

Qwen3.8-27Bとは?性能・必要VRAM・ローカル実行・商用利用を解説

Qwen3.8-27Bの性能、262K長文脈、必要VRAM、ローカル実行、商用利用を公式情報から解説

State of LLMs, July 2026: frontier, open-weight, and what actually fits on your own hardware

Frontier vs. open-weight performance matrix, plus what actually fits a 16GB local device

Your idle GPU box is a tax move and a private inference cloud at the same time

US — Section 179 / bonus depreciation timing, plus what the hardware can do the rest of the time

An inference cloud that can show its work

EU — what an audit-able run ledger looks like, for AI Act traceability

How murakumo.cloud's decentralization actually works today

Claim-by-claim: what's live (placement, CIDv1 tenant graphs, did:key/CACAO fleet auth) vs. roadmap (on-chain settlement, a token)