ceos/unsloth
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

fix(studio): prioritize curated defaults over HF download ranking in Recommended (#4792)

Browse source 
The model list merge order was `top_gguf + top_hub + static_models`,
which meant the HF download-ranked models always came first. New models
like Gemma 4 have low download counts and were not in the HF top-40,
so they got buried after 80 other models despite being at the top of
the curated static defaults in defaults.py.

Flip the merge to `static_models + top_gguf + top_hub` so editorial
picks (new model launches, promoted models) always appear first in the
Recommended section, with HF popularity backfilling after.

Co-authored-by: Daniel Han <danielhanchen@users.noreply.github.com>
This commit is contained in:
Daniel Han 2026-04-02 10:46:53 -07:00 • committed by GitHub
commit 7023e2a4ff
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
1 changed files with 3 additions and 2 deletions
Download patch file Download diff file Expand all files Collapse all files

5
studio/backend/core/inference/orchestrator.py
View file

@ -109,12 +109,13 @@ class InferenceOrchestrator:
self._top_models_ready.wait(timeout = 5)
top_gguf = self._top_gguf_cache or []
top_hub = self._top_hub_cache or []
# GGUFs first, then hub models, then static fallbacks.
# Curated static defaults first (editorial picks like new models),
# then HF download-ranked models to backfill.
# Send extras so the frontend still has 4 per category
# after removing already-downloaded models.
result: list[str] = []
seen: set[str] = set()
for m in top_gguf + top_hub + self._static_models:
for m in self._static_models + top_gguf + top_hub:
if m not in seen:
result.append(m)
seen.add(m)