Skip to content

Update Muse Glimmer GGUF filenames to canonical names #21853

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
sid-rp merged 1 commit into from
Aug 14, 2026
+15 −15
Merged

Update Muse Glimmer GGUF filenames to canonical names #21853

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
8 changes: 4 additions & 4 deletions .ci/scripts/export_model_artifact.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -525,10 +525,10 @@ if [ "$MODEL_NAME" = "muse_glimmer" ]; then

case "$QUANT_NAME" in
kquant-17gb)
TARGET_GGUF_FILE="muse-glimmer-30B-kquant-17gb.gguf"
TARGET_GGUF_FILE="Muse-Glimmer-30B-KQuant-17GB-Q4_K_M.gguf"
;;
kquant-dynamic)
TARGET_GGUF_FILE="muse-glimmer-30B-kquant-dynamic.gguf"
TARGET_GGUF_FILE="Muse-Glimmer-30B-KQuant-Dynamic-Q4_K_XL.gguf"
;;
esac

Expand All @@ -547,8 +547,8 @@ if [ "$MODEL_NAME" = "muse_glimmer" ]; then
--output-dir "${OUTPUT_DIR}"
;;
dflash-image)
DRAFT_GGUF_FILE="dflash-kquant.gguf"
MMPROJ_GGUF_FILE="mmproj-kquant.gguf"
DRAFT_GGUF_FILE="dflash-Muse-Glimmer-30B-Q4_K_M.gguf"
MMPROJ_GGUF_FILE="mmproj-Muse-Glimmer-30B-Q4_K_M.gguf"
python -c "from huggingface_hub import hf_hub_download; hf_hub_download('${HF_MODEL}', '${DRAFT_GGUF_FILE}', local_dir='${LOCAL_MODEL_DIR}')"
python -c "from huggingface_hub import hf_hub_download; hf_hub_download('${HF_MODEL}', '${MMPROJ_GGUF_FILE}', local_dir='${LOCAL_MODEL_DIR}')"
EXPORT_START_SECONDS=$SECONDS
Expand Down
14 changes: 7 additions & 7 deletions examples/models/muse-glimmer/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -44,17 +44,17 @@ the serving path uses it to render prompts and tools.

| File | Use |
|---|---|
| `muse-glimmer-30B-kquant-17gb.gguf` | Recommended target checkpoint |
| `muse-glimmer-30B-kquant-dynamic.gguf` | Dynamic K-quant target checkpoint |
| `mmproj-kquant.gguf` | Optional vision projector |
| `dflash-kquant.gguf` | Optional DFlash draft checkpoint |
| `Muse-Glimmer-30B-KQuant-17GB-Q4_K_M.gguf` | Recommended target checkpoint |
| `Muse-Glimmer-30B-KQuant-Dynamic-Q4_K_XL.gguf` | Dynamic K-quant target checkpoint |
| `mmproj-Muse-Glimmer-30B-Q4_K_M.gguf` | Optional vision projector |
| `dflash-Muse-Glimmer-30B-Q4_K_M.gguf` | Optional DFlash draft checkpoint |

Set paths once for the commands below:

```bash
TARGET="$(find assets/quant -type f -name 'muse-glimmer-30B-kquant-17gb.gguf' -print -quit)"
DRAFT="$(find assets/quant -type f -name 'dflash-kquant.gguf' -print -quit)"
MMPROJ="$(find assets/quant -type f -name 'mmproj-kquant.gguf' -print -quit)"
TARGET="$(find assets/quant -type f -name 'Muse-Glimmer-30B-KQuant-17GB-Q4_K_M.gguf' -print -quit)"
DRAFT="$(find assets/quant -type f -name 'dflash-Muse-Glimmer-30B-Q4_K_M.gguf' -print -quit)"
MMPROJ="$(find assets/quant -type f -name 'mmproj-Muse-Glimmer-30B-Q4_K_M.gguf' -print -quit)"
BACKEND=cuda # use mlx on macOS
```

Expand Down
4 changes: 2 additions & 2 deletions examples/models/muse-glimmer/export/export_dflash.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -716,7 +716,7 @@ def main() -> None:
target_src.add_argument(
"--target-gguf",
default=None,
help="Path to target GGUF (e.g. muse-glimmer-30B-kquant-17gb.gguf).",
help="Path to target GGUF (e.g. Muse-Glimmer-30B-KQuant-17GB-Q4_K_M.gguf).",
)
target_src.add_argument(
"--target-mlx",
Expand All @@ -728,7 +728,7 @@ def main() -> None:
draft_src.add_argument(
"--draft-gguf",
default=None,
help="Path to draft DFlash GGUF (e.g. dflash-kquant.gguf).",
help="Path to draft DFlash GGUF (e.g. dflash-Muse-Glimmer-30B-Q4_K_M.gguf).",
)
draft_src.add_argument(
"--draft-mlx",
Expand Down
4 changes: 2 additions & 2 deletions examples/models/muse-glimmer/export/export_solo.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -526,7 +526,7 @@ def main() -> None:
src.add_argument(
"--gguf",
default=None,
help="Path to a GGUF file (e.g. muse-glimmer-30B-kquant-17gb.gguf). Streams and "
help="Path to a GGUF file (e.g. Muse-Glimmer-30B-KQuant-17GB-Q4_K_M.gguf). Streams and "
"converts weights for the target backend, preserving quant bit-widths.",
)
src.add_argument(
Expand Down Expand Up @@ -592,7 +592,7 @@ def main() -> None:
"--mmproj",
default=None,
help="Optional path to the vision projector GGUF "
"(e.g. mmproj-kquant.gguf). When given (CUDA or "
"(e.g. mmproj-Muse-Glimmer-30B-Q4_K_M.gguf). When given (CUDA or "
"MLX backend), a `vision_encoder` method is exported and the runner can "
"take image input. Omit to export the text-only (TITO) model.",
)
Expand Down
Loading