ON THE RADAR

Ollama v0.40.6 launches clef in the cloud and adds GGUF q2_0

The October 16 release lets the clef decision model launch through Ollama's cloud, adds the q2_0 tensor and file types to GGUF support, and prevents orphaned runners on Windows.

Official source published Oct 16, 2026

Ollama released v0.40.6 on October 16, 2026, promoting the release candidate to stable. The patch-sized update contains four changes and no listed breaking changes. Read the release notes.

What changed

  • Launch clef in Ollama's cloud: the clef decision model can now be launched through the cloud, extending the cloud decision-model support introduced with clef:cloud in v0.40.5.
  • GGUF q2_0: the q2_0 tensor and file types were added to GGUF support.
  • Windows: a fix prevents orphaned runners.
  • Decision mode: --ask mode question keys now use identifiers.

This report covers the October 16 release notes; earlier releases are covered in our reports on v0.40.5, v0.40.4 and v0.40.3.

Original sourceRead the official source