feat(models): Muse-Glimmer-30B mit DFlash-Speculative-Drafting in Katalog und llama-swap integriert
This commit is contained in:
@@ -32,6 +32,17 @@ models:
|
||||
out: [text]
|
||||
tools: true
|
||||
context: 65536
|
||||
Muse-Glimmer-30B:
|
||||
# Meta Muse Glimmer 30B: Dichtes 30B-Agenten-Modell mit DFlash-Speculative-Drafting
|
||||
# und 1.8B Perception Multimodal Projector.
|
||||
cmd: |
|
||||
llama-server -m /srv/models/Muse-Glimmer-30B-GGUF/Muse-Glimmer-30B-UD-Q4_K_XL.gguf --host 127.0.0.1 --port ${PORT} -c 65536 -ngl 999 -fa on --no-mmap --mmproj /srv/models/Muse-Glimmer-30B-GGUF/mmproj-kquant.gguf --jinja --spec-type draft-dflash --spec-draft-model /srv/models/Muse-Glimmer-30B-GGUF/dflash-kquant.gguf
|
||||
ttl: 600
|
||||
capabilities:
|
||||
in: [text, image]
|
||||
out: [text]
|
||||
tools: true
|
||||
context: 65536
|
||||
Qwen3-Coder-Next:
|
||||
cmd: |
|
||||
llama-server -m /srv/models/Qwen3-Coder-Next-GGUF/Qwen3-Coder-Next-Q4_K_M.gguf --host 127.0.0.1 --port ${PORT} -c 131072 -ngl 999 -fa on --no-mmap --jinja --cache-reuse 256 -cram 16384
|
||||
|
||||
Reference in New Issue
Block a user