mirror of
https://github.com/mudler/LocalAI.git
synced 2026-09-12 22:33:54 -04:00
feat(gallery): add DFM Mimir for vLLM
DFM Mimir is a new permissively licensed Danish and English HRM-Text model. LocalAI already ships a vLLM version with native support for its architecture. Assisted-by: Codex:gpt-5 Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
This commit is contained in:
1 parent
6324ee1dd9
commit
c96904ddf4
1 file changed
+29
@@ -393,6 +393,35 @@
|
||||
- embeddings
|
||||
parameters:
|
||||
model: tencent/WeMM-Embedding-9B
|
||||
- name: "dfm-mimir:vllm"
|
||||
url: github:mudler/LocalAI/gallery/vllm.yaml@master
|
||||
urls:
|
||||
- https://huggingface.co/danish-foundation-models/DFM-Mimir
|
||||
description: |
|
||||
DFM Mimir is an Apache-2.0, instruction-tuned HRM-Text model from Danish
|
||||
Foundation Models. It has about 1 billion parameters and a 4,096-token
|
||||
context window. The model focuses on Danish and English chat, reasoning,
|
||||
mathematics, and code generation, and uses only permissible post-training
|
||||
data. This entry serves the official BF16 safetensors checkpoint with vLLM.
|
||||
license: apache-2.0
|
||||
tags:
|
||||
- llm
|
||||
- chat
|
||||
- reasoning
|
||||
- coding
|
||||
- instruction-tuned
|
||||
- danish
|
||||
- english
|
||||
- hrm-text
|
||||
- safetensors
|
||||
- vllm
|
||||
- gpu
|
||||
last_checked: "2026-08-25"
|
||||
overrides:
|
||||
context_size: 4096
|
||||
parameters:
|
||||
max_tokens: 4096
|
||||
model: danish-foundation-models/DFM-Mimir
|
||||
- &granite-4-2-3b
|
||||
name: "granite-4.2-3b-q4"
|
||||
variants:
|
||||
|
||||
Reference in new issue
Block a user