Skip to content

feat(gallery): add Qwen3.8 Cold Fusion - #11754

Open
localai-org-maint-bot wants to merge 1 commit into
masterfrom
cron/model-gallery-20260827-run2
Open

feat(gallery): add Qwen3.8 Cold Fusion#11754
localai-org-maint-bot wants to merge 1 commit into
masterfrom
cron/model-gallery-20260827-run2

Conversation

@localai-org-maint-bot

Copy link
Copy Markdown
Collaborator

Description

Adds Qwen3.8 27B Cold Fusion to the model gallery with Q4_K_M and Q8_0 MTP-enabled GGUF variants. Both variants include the shared BF16 vision projector and use the publisher-recommended thinking-mode sampling defaults.

The model is a recent Apache-2.0 Qwen3.8 fine-tune with approximately 239k Hugging Face downloads at discovery time.

Notes for Reviewers

  • Verified all three SHA256 values against Hugging Face x-linked-etag headers.
  • git diff --check passes.
  • go test ./core/gallery could not start because the required generated protobuf package is absent. make protogen-go was blocked by a 403 while downloading protoc through the environment proxy.
  • The new entries use the existing Qwen3.8 llama.cpp support.

Signed commits

  • Yes, I signed my commits.
  • Documentation updated (docs/content/) for user-facing changes, or not applicable

Add Q4_K_M and Q8_0 MTP variants with the shared vision projector.
The publisher recommends these builds for faster Qwen3.8 generation.

Assisted-by: Codex:gpt-5
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