Skip to content

0863: catalog offers the installed engine first; GGUF builds for 39 curated chat rows - #13

Merged
lpalbou merged 4 commits into
mainfrom
fix/catalog-installed-engines-first
Oct 29, 2026
Merged

lpalbou merged 4 commits into
mainfrom
fix/catalog-installed-engines-first

Conversation

@lpalbou

@lpalbou lpalbou commented Oct 29, 2026

Copy link
Copy Markdown
Owner

Backlog 0863 (OVH server QA, front item 3). On a llama.cpp-only Linux host, the Qwen3.5 9B row listed and pre-selected only LM Studio/Ollama builds, and the console showed "Hugging Face 0".

  • Seed: 39 GGUF artifacts (Q4_K_M; gpt-oss-120b MXFP4). Repo id, size (files the *QUANT*.gguf pattern selects) and revision were each read from the HF API on 2026-10-29. Qwen3.8 Flash-Next is skipped: the repo has no plain Q4_K_M.
  • _pick_recommended: a curated pick for an absent engine yields to an artifact an installed engine serves. The fallback rank puts an installed engine before the fit verdict, unless the artifact is too_large. Apple tiers and forced fallbacks are unchanged.
  • _order_for_installed_engines: each artifact carries engine_installed; a row lists the pre-selection first, then installed engines' artifacts.
  • Tests: llama.cpp-only host (mutation-checked: red without the yield, red without the sort); coverage of every chat row; the starter test now runs with no engine installed. tests/config: 1025 passed. Docs (models.md) and CHANGELOG [Unreleased] updated.

No version bump.

@lpalbou
lpalbou merged commit 79fdf1e into main Oct 29, 2026
11 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant