feat(gallery): add Qwen3.8 GSQ-RCO variants - #11787
Merged
Merged
Conversation
Add three llama.cpp-compatible mixed quantizations from ISTA DASLab. These builds give Qwen3.8-27B users an 8.4 to 10.1 GB weight tier with the shared vision projector. Assisted-by: Codex:gpt-5
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Description
Add the three Qwen3.8-27B GSQ-RCO GGUF quantizations published by ISTA DASLab. The new variants provide 8.4 GB, 9.3 GB, and 10.1 GB weight options, and each uses the shared BF16 vision projector.
The entries use the existing Qwen3.8-27B llama.cpp configuration and are linked from its variant list. Every artifact SHA-256 matches the Hugging Face x-linked-etag.
Verification
go test ./core/gallery -ginkgo.focus="gallery/index.yaml"git diff --checkThe full
go test ./core/galleryrun reached 383 passing specs, but three unrelated network-dependent specs failed because this environment receives HTTP 403 from raw.githubusercontent.com.Notes for Reviewers
The upstream model card identifies IQ3_XXS as the recommended highest-quality build in this set. Automatic variant resolution still ranks all same-engine builds by measured size and fit.
Signed commits
The commit is intentionally unsigned because AI assistants cannot provide DCO sign-off.