models(gallery): add gemma-2b

gallery/gemma.yaml

@@ -0,0 +1,20 @@
+---
+name: "gemma"
+
+config_file: |
+  mmap: true
+  context_size: 8192
+  template:
+    chat_message: |-
+      <start_of_turn>{{if eq .RoleName "assistant" }}model{{else}}{{ .RoleName }}{{end}}
+      {{ if .Content -}}
+      {{.Content }}
+      {{ end -}}<end_of_turn>
+    chat: |
+      {{.Input -}}
+      <start_of_turn>model
+    completion: |
+      {{.Input}}
+  stopwords:
+  - '<|im_end|>'
+  - '<end_of_turn>'

gallery/index.yaml

@@ -131,6 +131,28 @@
     - filename: Einstein-v6.1-Llama3-8B-Q4_K_M.gguf
       sha256: 447587bd8f60d9050232148d34fdb2d88b15b2413fd7f8e095a4606ec60b45bf
       uri: huggingface://bartowski/Einstein-v6.1-Llama3-8B-GGUF/Einstein-v6.1-Llama3-8B-Q4_K_M.gguf
+- &gemma
+  url: "github:mudler/LocalAI/gallery/gemma.yaml@master"
+  name: "gemma-2b"
+  license: gemma
+  urls:
+    - https://ai.google.dev/gemma/docs
+    - https://huggingface.co/mlabonne/gemma-2b-GGUF
+  description: |
+    Open source LLM from Google
+  tags:
+    - llm
+    - gguf
+    - gpu
+    - cpu
+    - gemma
+  overrides:
+    parameters:
+      model: gemma-2b.Q4_K_M.gguf
+  files:
+    - filename: gemma-2b.Q4_K_M.gguf
+      sha256: 37d50c21ef7847926204ad9b3007127d9a2722188cfd240ce7f9f7f041aa71a5
+      uri: huggingface://mlabonne/gemma-2b-GGUF/gemma-2b.Q4_K_M.gguf
 - &llama3
   url: "github:mudler/LocalAI/gallery/llama3-instruct.yaml@master"
   icon: https://cdn-uploads.huggingface.co/production/uploads/642cc1c253e76b4c2286c58e/aJJxKus1wP5N-euvHEUq7.png