feat(gallery): add Sage-A1.5 GGUF variants

docs/content/features/model-gallery.md

@@ -60,6 +60,20 @@ local-ai models install nemotron-3.5-bolt-e4b-q4
 To select a build explicitly, use `--variant nemotron-3.5-bolt-e4b-q4`
 or `--variant nemotron-3.5-bolt-e4b-q8`.
 
+### Sage-A1.5
+
+[Sage-A1.5](https://huggingface.co/InternScience/Sage-A1.5) supports text chat
+and image understanding through llama.cpp. The gallery includes the publisher's
+Q4_K_M and Q8_0 GGUF builds with the vision projector and a 32,768-token default context.
+
+Install with automatic variant selection:
+
+```bash
+local-ai models install sage-a1.5-q4
+```
+
+Use `--variant sage-a1.5-q4` or `--variant sage-a1.5-q8` to select a specific build.
+
 ### MiniCPM-V 4.7 1B
 
 [MiniCPM-V 4.7 1B](https://huggingface.co/openbmb/MiniCPM-V-4.7-1B)

gallery/index.yaml

@@ -3,6 +3,88 @@
 # written by scripts/models/cmd/modelcatalog and stays only so releases before
 # the cutover keep working. CI fails when it differs from the generated copy:
 #   go run ./scripts/models/cmd/modelcatalog --library gallery/models/index.yaml --out gallery/index.yaml
+- url: github:mudler/LocalAI/gallery/virtual.yaml@master
+  name: sage-a1.5-q4
+  description: |
+    Sage-A1.5 is a 35B mixture-of-experts model based on Qwen3.6 for
+    scientific reasoning, coding, and tool use. This official Q4_K_M GGUF
+    build uses llama.cpp and includes the vision projector.
+  license: apache-2.0
+  urls:
+    - https://huggingface.co/InternScience/Sage-A1.5
+    - https://huggingface.co/InternScience/Sage-A1.5-Q4_K_M-GGUF
+  tags:
+    - llm
+    - gguf
+    - vision
+    - reasoning
+    - cpu
+    - gpu
+  files:
+    - filename: Sage-A1.5-Q4_K_M.gguf
+      sha256: a64fe0bbf936db3eb54f66f142c2b2eee569ade83673649f549a21daef8ca395
+      uri: https://huggingface.co/InternScience/Sage-A1.5-Q4_K_M-GGUF/resolve/10b179b52f2c841030db327627e707d076b4c8cf/Sage-A1.5-Q4_K_M.gguf
+    - filename: Sage-A1.5-mmproj.gguf
+      sha256: 51cd4410057e8dc4152ca771a726fbfb34856f047e7049018c757d42ddd481f2
+      uri: https://huggingface.co/InternScience/Sage-A1.5-Q4_K_M-GGUF/resolve/10b179b52f2c841030db327627e707d076b4c8cf/Sage-A1.5-mmproj.gguf
+  overrides:
+    backend: llama-cpp
+    context_size: 32768
+    known_usecases:
+      - chat
+      - vision
+    mmproj: Sage-A1.5-mmproj.gguf
+    options:
+      - use_jinja:true
+    parameters:
+      model: Sage-A1.5-Q4_K_M.gguf
+      temperature: 1.0
+      top_k: 20
+      top_p: 0.95
+    template:
+      use_tokenizer_template: true
+  variants:
+    - model: sage-a1.5-q8
+- url: github:mudler/LocalAI/gallery/virtual.yaml@master
+  name: sage-a1.5-q8
+  description: |
+    Sage-A1.5 is a 35B mixture-of-experts model based on Qwen3.6 for
+    scientific reasoning, coding, and tool use. This official Q8_0 GGUF
+    build uses llama.cpp and includes the vision projector.
+  license: apache-2.0
+  urls:
+    - https://huggingface.co/InternScience/Sage-A1.5
+    - https://huggingface.co/InternScience/Sage-A1.5-Q8_0-GGUF
+  tags:
+    - llm
+    - gguf
+    - vision
+    - reasoning
+    - cpu
+    - gpu
+  files:
+    - filename: Sage-A1.5-Q8_0.gguf
+      sha256: f7247f6faef3f6c616b953595e1fc7c3fbf257d02d9915af5fdf15e808ea8ee4
+      uri: https://huggingface.co/InternScience/Sage-A1.5-Q8_0-GGUF/resolve/1cbe2925983a13a6c09881b9bc088884dbfbc146/Sage-A1.5-Q8_0.gguf
+    - filename: Sage-A1.5-mmproj.gguf
+      sha256: 51cd4410057e8dc4152ca771a726fbfb34856f047e7049018c757d42ddd481f2
+      uri: https://huggingface.co/InternScience/Sage-A1.5-Q4_K_M-GGUF/resolve/10b179b52f2c841030db327627e707d076b4c8cf/Sage-A1.5-mmproj.gguf
+  overrides:
+    backend: llama-cpp
+    context_size: 32768
+    known_usecases:
+      - chat
+      - vision
+    mmproj: Sage-A1.5-mmproj.gguf
+    options:
+      - use_jinja:true
+    parameters:
+      model: Sage-A1.5-Q8_0.gguf
+      temperature: 1.0
+      top_k: 20
+      top_p: 0.95
+    template:
+      use_tokenizer_template: true
 - url: github:mudler/LocalAI/gallery/virtual.yaml@master
   name: adavar-3b-q4
   description: |

gallery/models/index.yaml

@@ -4,6 +4,104 @@
 # changes and never edit it by hand. CI fails when it is stale. A build from
 # master reads it as the default model source; a tagged build reads the gallery
 # until the release cutover.
+- kind: model
+  id: sage-a1.5-q4
+  name: sage-a1.5-q4
+  version: 0.0.0-gallery.88087a1aab17
+  description: |
+    Sage-A1.5 is a 35B mixture-of-experts model based on Qwen3.6 for
+    scientific reasoning, coding, and tool use. This official Q4_K_M GGUF
+    build uses llama.cpp and includes the vision projector.
+  license: apache-2.0
+  urls:
+    - https://huggingface.co/InternScience/Sage-A1.5
+    - https://huggingface.co/InternScience/Sage-A1.5-Q4_K_M-GGUF
+  tags:
+    - llm
+    - gguf
+    - vision
+    - reasoning
+    - cpu
+    - gpu
+  channel: gallery
+  policy:
+    requireSignature: false
+  schemaVersion: 1
+  model:
+    url: github:mudler/LocalAI/gallery/virtual.yaml@master
+    files:
+      - filename: Sage-A1.5-Q4_K_M.gguf
+        sha256: a64fe0bbf936db3eb54f66f142c2b2eee569ade83673649f549a21daef8ca395
+        uri: https://huggingface.co/InternScience/Sage-A1.5-Q4_K_M-GGUF/resolve/10b179b52f2c841030db327627e707d076b4c8cf/Sage-A1.5-Q4_K_M.gguf
+      - filename: Sage-A1.5-mmproj.gguf
+        sha256: 51cd4410057e8dc4152ca771a726fbfb34856f047e7049018c757d42ddd481f2
+        uri: https://huggingface.co/InternScience/Sage-A1.5-Q4_K_M-GGUF/resolve/10b179b52f2c841030db327627e707d076b4c8cf/Sage-A1.5-mmproj.gguf
+    overrides:
+      backend: llama-cpp
+      context_size: 32768
+      known_usecases:
+        - chat
+        - vision
+      mmproj: Sage-A1.5-mmproj.gguf
+      options:
+        - use_jinja:true
+      parameters:
+        model: Sage-A1.5-Q4_K_M.gguf
+        temperature: 1.0
+        top_k: 20
+        top_p: 0.95
+      template:
+        use_tokenizer_template: true
+    variants:
+      - model: sage-a1.5-q8
+- kind: model
+  id: sage-a1.5-q8
+  name: sage-a1.5-q8
+  version: 0.0.0-gallery.d2ee65a3043e
+  description: |
+    Sage-A1.5 is a 35B mixture-of-experts model based on Qwen3.6 for
+    scientific reasoning, coding, and tool use. This official Q8_0 GGUF
+    build uses llama.cpp and includes the vision projector.
+  license: apache-2.0
+  urls:
+    - https://huggingface.co/InternScience/Sage-A1.5
+    - https://huggingface.co/InternScience/Sage-A1.5-Q8_0-GGUF
+  tags:
+    - llm
+    - gguf
+    - vision
+    - reasoning
+    - cpu
+    - gpu
+  channel: gallery
+  policy:
+    requireSignature: false
+  schemaVersion: 1
+  model:
+    url: github:mudler/LocalAI/gallery/virtual.yaml@master
+    files:
+      - filename: Sage-A1.5-Q8_0.gguf
+        sha256: f7247f6faef3f6c616b953595e1fc7c3fbf257d02d9915af5fdf15e808ea8ee4
+        uri: https://huggingface.co/InternScience/Sage-A1.5-Q8_0-GGUF/resolve/1cbe2925983a13a6c09881b9bc088884dbfbc146/Sage-A1.5-Q8_0.gguf
+      - filename: Sage-A1.5-mmproj.gguf
+        sha256: 51cd4410057e8dc4152ca771a726fbfb34856f047e7049018c757d42ddd481f2
+        uri: https://huggingface.co/InternScience/Sage-A1.5-Q4_K_M-GGUF/resolve/10b179b52f2c841030db327627e707d076b4c8cf/Sage-A1.5-mmproj.gguf
+    overrides:
+      backend: llama-cpp
+      context_size: 32768
+      known_usecases:
+        - chat
+        - vision
+      mmproj: Sage-A1.5-mmproj.gguf
+      options:
+        - use_jinja:true
+      parameters:
+        model: Sage-A1.5-Q8_0.gguf
+        temperature: 1.0
+        top_k: 20
+        top_p: 0.95
+      template:
+        use_tokenizer_template: true
 - kind: model
   id: adavar-3b-q4
   name: adavar-3b-q4