Skip to content

v0.8.0 — value score + homebench doctor

Choose a tag to compare

View all tags
@david-g-3654 david-g-3654 released this 06 Aug 00:53
· 9 commits to main since this release
v0.8.0
b2f5953
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.

"Best model for your laptop" value score + homebench doctor

  • 🏆 Value score — a composite 0–100 score blending quality, tok/s, and memory (normalised within your run), so homebench names the best model for your hardware, not just a ranking. Prints a verdict after multi-model runs and adds a Value column + callout to the Markdown/HTML reports.
  • homebench doctor — one command to diagnose setup: Python, hardware + model budget, reachable providers + pulled models, home/cache writability, optional pyyaml, and installed-vs-PyPI version (with a proper version compare). Exits non-zero on hard failures.

Both were roadmap items.

Full changelog: v0.7.0...v0.8.0

Assets 2
Loading