freenode

IRC About

Kernel Distros Desktop Internet Web Languages Infra Security Cryptocurrency AI Policy Self-Hosted Digests

#threading

llama.cpp auto thread count now prefers physical cores

When users pass -t -1, the runtime sizes worker threads to physical cores rather than SMT logical cores.

Sections

Kernel & Low-Level Distributions & Plumbing Desktop & Graphics Internet & Protocols Web Platform Languages & Toolchains Databases & Infrastructure Security & Cryptography Cryptocurrency AI & ML Policy & Civil Liberties Self-Hosted

Read

Daily digests Releases Security advisories RSS

Company

About IRC Network Write for us Contact Send a tip

Legal

Terms of Use Privacy Policy

© 2026 freenode. All rights reserved.

Product, project, and company names mentioned on this site are trademarks of their respective owners, and are referenced solely for purposes of identification and news reporting. Material on this site is provided for general informational purposes only, is subject to change without notice, and is provided without warranty of any kind. Nothing on this site constitutes professional, legal, financial, or investment advice, an offer or solicitation of any kind, or a recommendation with respect to any digital asset. Reporting draws on publicly available primary sources; quoted material remains the property of its authors. Use of this site is subject to the Terms of Use and Privacy Policy.