feat(gallery): add Nemotron Bolt E4B variants (#12880)

Add Q4 and Q8 GGUF builds with the F16 vision projector. Link the builds
as variants and document installation.

Assisted-by: Codex:gpt-6

Co-authored-by: localai-org-maint-bot <306269227+localai-org-maint-bot@users.noreply.github.com>

docs/content/features/model-gallery.md

@@ -44,6 +44,22 @@ Both views store the view, search, filter, and selection in the URL. Installing
 from Explore does not move you away from the catalog; the entry updates in place
 when the operation finishes.
 
+### Nemotron 3.5 Bolt E4B
+
+[Nemotron 3.5 Bolt E4B](https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Bolt-E4B-BF16)
+is NVIDIA's Gemma 4 E4B fine-tune for conversational assistants and tool use.
+The llama.cpp entries include Unsloth's UD-Q4_K_XL and UD-Q8_K_XL builds,
+the F16 vision projector, and a 32,768-token default context.
+
+Install with automatic variant selection:
+
+```bash
+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`.
+
 ### MiniCPM-V 4.7 1B
 
 [MiniCPM-V 4.7 1B](https://huggingface.co/openbmb/MiniCPM-V-4.7-1B)

gallery/index.yaml

@@ -16891,6 +16891,100 @@
       top_p: 0.95
     template:
       use_tokenizer_template: true
+- url: github:mudler/LocalAI/gallery/virtual.yaml@master
+  name: nemotron-3.5-bolt-e4b-q4
+  description: |
+    NVIDIA Nemotron 3.5 Bolt E4B is a Gemma 4 E4B fine-tune for
+    conversational assistants, instruction following, and multi-turn tool use.
+    This UD-Q4_K_XL GGUF build includes the F16 vision projector for
+    chat and image input, with a 32,768-token context by default.
+  license: openmdw-1.1
+  urls:
+    - https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Bolt-E4B-BF16
+    - https://huggingface.co/unsloth/NVIDIA-Nemotron-3.5-Bolt-E4B-GGUF
+  tags:
+    - llm
+    - gguf
+    - cpu
+    - gpu
+    - gemma4
+    - multimodal
+    - reasoning
+  files:
+    - filename: llama-cpp/models/nemotron-3.5-bolt-e4b/NVIDIA-Nemotron-3.5-Bolt-E4B-UD-Q4_K_XL.gguf
+      sha256: 84ff34f5834b929043f5ae74799e8d562d4056998f6a5923ceb44831e8ad84d6
+      uri: https://huggingface.co/unsloth/NVIDIA-Nemotron-3.5-Bolt-E4B-GGUF/resolve/b738967d6da3441a7d1c71fcec15e49e32e16ecd/NVIDIA-Nemotron-3.5-Bolt-E4B-UD-Q4_K_XL.gguf
+    - filename: llama-cpp/mmproj/nemotron-3.5-bolt-e4b/mmproj-F16.gguf
+      sha256: 6bc77b7ac2b62cf61bc86757605e45e39f6345373f0a6b76cc5b1ed97f2e233e
+      uri: https://huggingface.co/unsloth/NVIDIA-Nemotron-3.5-Bolt-E4B-GGUF/resolve/b738967d6da3441a7d1c71fcec15e49e32e16ecd/mmproj-F16.gguf
+  overrides:
+    backend: llama-cpp
+    context_size: 32768
+    function:
+      automatic_tool_parsing_fallback: true
+      grammar:
+        disable: true
+    known_usecases:
+      - chat
+      - vision
+    mmproj: llama-cpp/mmproj/nemotron-3.5-bolt-e4b/mmproj-F16.gguf
+    options:
+      - use_jinja:true
+    parameters:
+      model: llama-cpp/models/nemotron-3.5-bolt-e4b/NVIDIA-Nemotron-3.5-Bolt-E4B-UD-Q4_K_XL.gguf
+      temperature: 1
+      top_k: 64
+      top_p: 0.95
+    template:
+      use_tokenizer_template: true
+  variants:
+    - model: nemotron-3.5-bolt-e4b-q8
+- url: github:mudler/LocalAI/gallery/virtual.yaml@master
+  name: nemotron-3.5-bolt-e4b-q8
+  description: |
+    NVIDIA Nemotron 3.5 Bolt E4B is a Gemma 4 E4B fine-tune for
+    conversational assistants, instruction following, and multi-turn tool use.
+    This UD-Q8_K_XL GGUF build includes the F16 vision projector for
+    chat and image input, with a 32,768-token context by default.
+  license: openmdw-1.1
+  urls:
+    - https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Bolt-E4B-BF16
+    - https://huggingface.co/unsloth/NVIDIA-Nemotron-3.5-Bolt-E4B-GGUF
+  tags:
+    - llm
+    - gguf
+    - cpu
+    - gpu
+    - gemma4
+    - multimodal
+    - reasoning
+  files:
+    - filename: llama-cpp/models/nemotron-3.5-bolt-e4b/NVIDIA-Nemotron-3.5-Bolt-E4B-UD-Q8_K_XL.gguf
+      sha256: 6176bd788769bd58042c165d3dc8e8c86a8ca5c90d5de188a364b31dc9560810
+      uri: https://huggingface.co/unsloth/NVIDIA-Nemotron-3.5-Bolt-E4B-GGUF/resolve/b738967d6da3441a7d1c71fcec15e49e32e16ecd/NVIDIA-Nemotron-3.5-Bolt-E4B-UD-Q8_K_XL.gguf
+    - filename: llama-cpp/mmproj/nemotron-3.5-bolt-e4b/mmproj-F16.gguf
+      sha256: 6bc77b7ac2b62cf61bc86757605e45e39f6345373f0a6b76cc5b1ed97f2e233e
+      uri: https://huggingface.co/unsloth/NVIDIA-Nemotron-3.5-Bolt-E4B-GGUF/resolve/b738967d6da3441a7d1c71fcec15e49e32e16ecd/mmproj-F16.gguf
+  overrides:
+    backend: llama-cpp
+    context_size: 32768
+    function:
+      automatic_tool_parsing_fallback: true
+      grammar:
+        disable: true
+    known_usecases:
+      - chat
+      - vision
+    mmproj: llama-cpp/mmproj/nemotron-3.5-bolt-e4b/mmproj-F16.gguf
+    options:
+      - use_jinja:true
+    parameters:
+      model: llama-cpp/models/nemotron-3.5-bolt-e4b/NVIDIA-Nemotron-3.5-Bolt-E4B-UD-Q8_K_XL.gguf
+      temperature: 1
+      top_k: 64
+      top_p: 0.95
+    template:
+      use_tokenizer_template: true
 - url: github:mudler/LocalAI/gallery/virtual.yaml@master
   name: gemma-4-e4b-heretic-q4
   description: |

gallery/models/index.yaml

@@ -19708,6 +19708,116 @@
         top_p: 0.95
       template:
         use_tokenizer_template: true
+- kind: model
+  id: nemotron-3.5-bolt-e4b-q4
+  name: nemotron-3.5-bolt-e4b-q4
+  version: 1.0.0
+  description: |
+    NVIDIA Nemotron 3.5 Bolt E4B is a Gemma 4 E4B fine-tune for
+    conversational assistants, instruction following, and multi-turn tool use.
+    This UD-Q4_K_XL GGUF build includes the F16 vision projector for
+    chat and image input, with a 32,768-token context by default.
+  license: openmdw-1.1
+  urls:
+    - https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Bolt-E4B-BF16
+    - https://huggingface.co/unsloth/NVIDIA-Nemotron-3.5-Bolt-E4B-GGUF
+  tags:
+    - llm
+    - gguf
+    - cpu
+    - gpu
+    - gemma4
+    - multimodal
+    - reasoning
+  channel: gallery
+  policy:
+    requireSignature: false
+  schemaVersion: 1
+  model:
+    url: github:mudler/LocalAI/gallery/virtual.yaml@master
+    files:
+      - filename: llama-cpp/models/nemotron-3.5-bolt-e4b/NVIDIA-Nemotron-3.5-Bolt-E4B-UD-Q4_K_XL.gguf
+        sha256: 84ff34f5834b929043f5ae74799e8d562d4056998f6a5923ceb44831e8ad84d6
+        uri: https://huggingface.co/unsloth/NVIDIA-Nemotron-3.5-Bolt-E4B-GGUF/resolve/b738967d6da3441a7d1c71fcec15e49e32e16ecd/NVIDIA-Nemotron-3.5-Bolt-E4B-UD-Q4_K_XL.gguf
+      - filename: llama-cpp/mmproj/nemotron-3.5-bolt-e4b/mmproj-F16.gguf
+        sha256: 6bc77b7ac2b62cf61bc86757605e45e39f6345373f0a6b76cc5b1ed97f2e233e
+        uri: https://huggingface.co/unsloth/NVIDIA-Nemotron-3.5-Bolt-E4B-GGUF/resolve/b738967d6da3441a7d1c71fcec15e49e32e16ecd/mmproj-F16.gguf
+    overrides:
+      backend: llama-cpp
+      context_size: 32768
+      function:
+        automatic_tool_parsing_fallback: true
+        grammar:
+          disable: true
+      known_usecases:
+        - chat
+        - vision
+      mmproj: llama-cpp/mmproj/nemotron-3.5-bolt-e4b/mmproj-F16.gguf
+      options:
+        - use_jinja:true
+      parameters:
+        model: llama-cpp/models/nemotron-3.5-bolt-e4b/NVIDIA-Nemotron-3.5-Bolt-E4B-UD-Q4_K_XL.gguf
+        temperature: 1
+        top_k: 64
+        top_p: 0.95
+      template:
+        use_tokenizer_template: true
+    variants:
+      - model: nemotron-3.5-bolt-e4b-q8
+- kind: model
+  id: nemotron-3.5-bolt-e4b-q8
+  name: nemotron-3.5-bolt-e4b-q8
+  version: 1.0.0
+  description: |
+    NVIDIA Nemotron 3.5 Bolt E4B is a Gemma 4 E4B fine-tune for
+    conversational assistants, instruction following, and multi-turn tool use.
+    This UD-Q8_K_XL GGUF build includes the F16 vision projector for
+    chat and image input, with a 32,768-token context by default.
+  license: openmdw-1.1
+  urls:
+    - https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Bolt-E4B-BF16
+    - https://huggingface.co/unsloth/NVIDIA-Nemotron-3.5-Bolt-E4B-GGUF
+  tags:
+    - llm
+    - gguf
+    - cpu
+    - gpu
+    - gemma4
+    - multimodal
+    - reasoning
+  channel: gallery
+  policy:
+    requireSignature: false
+  schemaVersion: 1
+  model:
+    url: github:mudler/LocalAI/gallery/virtual.yaml@master
+    files:
+      - filename: llama-cpp/models/nemotron-3.5-bolt-e4b/NVIDIA-Nemotron-3.5-Bolt-E4B-UD-Q8_K_XL.gguf
+        sha256: 6176bd788769bd58042c165d3dc8e8c86a8ca5c90d5de188a364b31dc9560810
+        uri: https://huggingface.co/unsloth/NVIDIA-Nemotron-3.5-Bolt-E4B-GGUF/resolve/b738967d6da3441a7d1c71fcec15e49e32e16ecd/NVIDIA-Nemotron-3.5-Bolt-E4B-UD-Q8_K_XL.gguf
+      - filename: llama-cpp/mmproj/nemotron-3.5-bolt-e4b/mmproj-F16.gguf
+        sha256: 6bc77b7ac2b62cf61bc86757605e45e39f6345373f0a6b76cc5b1ed97f2e233e
+        uri: https://huggingface.co/unsloth/NVIDIA-Nemotron-3.5-Bolt-E4B-GGUF/resolve/b738967d6da3441a7d1c71fcec15e49e32e16ecd/mmproj-F16.gguf
+    overrides:
+      backend: llama-cpp
+      context_size: 32768
+      function:
+        automatic_tool_parsing_fallback: true
+        grammar:
+          disable: true
+      known_usecases:
+        - chat
+        - vision
+      mmproj: llama-cpp/mmproj/nemotron-3.5-bolt-e4b/mmproj-F16.gguf
+      options:
+        - use_jinja:true
+      parameters:
+        model: llama-cpp/models/nemotron-3.5-bolt-e4b/NVIDIA-Nemotron-3.5-Bolt-E4B-UD-Q8_K_XL.gguf
+        temperature: 1
+        top_k: 64
+        top_p: 0.95
+      template:
+        use_tokenizer_template: true
 - kind: model
   id: gemma-4-e4b-heretic-q4
   name: gemma-4-e4b-heretic-q4