diff --git a/gallery/index.yaml b/gallery/index.yaml
index 0afca67fb3..f13993cbbc 100644
--- a/gallery/index.yaml
+++ b/gallery/index.yaml
@@ -1394,6 +1394,335 @@
       - pooling:cls
     parameters:
       model: bge-small-en-v1.5-q4_k_m.gguf
+- url: github:mudler/LocalAI/gallery/virtual.yaml@master
+  name: gemma-4-e2b-it-artemis-q3-k-l
+  description: |
+    TurinTech Artemis Q3_K_L quantization of Gemma 4 E2B from Google's QAT
+    checkpoint, for local chat and tool calls. Accepts image input through
+    the bundled F16 vision projector. Uses the embedded chat template and
+    an 8192-token context. Thinking is on by default.
+  license: apache-2.0
+  urls:
+    - https://huggingface.co/google/gemma-4-E2B-it-qat-q4_0-unquantized
+    - https://huggingface.co/turintech/gemma-4-E2B-it-Artemis-GGUF
+  icon: https://huggingface.co/turintech/gemma-4-E2B-it-Artemis-GGUF/resolve/13a1c266e43df6e184779f8bc446ca6c66dca152/assets/turintech-logo.png
+  tags:
+    - llm
+    - gguf
+    - cpu
+    - gpu
+    - gemma
+    - gemma4
+    - gemma-4
+    - multimodal
+    - vision
+  files:
+    - filename: llama-cpp/models/gemma-4-E2B-it-Artemis-GGUF/gemma-4-E2B-it-Artemis-Q3_K_L.gguf
+      sha256: 1e207a449e37ea902de237aba1158f960a91bd344bee6edc5b0aa0dbce015621
+      uri: https://huggingface.co/turintech/gemma-4-E2B-it-Artemis-GGUF/resolve/13a1c266e43df6e184779f8bc446ca6c66dca152/gemma-4-E2B-it-Artemis-Q3_K_L.gguf
+    - filename: llama-cpp/mmproj/gemma-4-E2B-it-Artemis-GGUF/mmproj-F16.gguf
+      sha256: 35be86a206b3573bd149e59b0ea8d8c371dab6f407e1b3f4f73d6408ff6ce6bf
+      uri: https://huggingface.co/turintech/gemma-4-E2B-it-Artemis-GGUF/resolve/13a1c266e43df6e184779f8bc446ca6c66dca152/mmproj-F16.gguf
+  overrides:
+    backend: llama-cpp
+    context_size: 8192
+    function:
+      automatic_tool_parsing_fallback: true
+      grammar:
+        disable: true
+    known_usecases:
+      - chat
+      - vision
+    mmproj: llama-cpp/mmproj/gemma-4-E2B-it-Artemis-GGUF/mmproj-F16.gguf
+    options:
+      - use_jinja:true
+    parameters:
+      model: llama-cpp/models/gemma-4-E2B-it-Artemis-GGUF/gemma-4-E2B-it-Artemis-Q3_K_L.gguf
+    template:
+      use_tokenizer_template: true
+  variants:
+    - model: gemma-4-e2b-it-artemis-q3-k-m
+    - model: gemma-4-e2b-it-artemis-iq3-xxs
+    - model: gemma-4-e2b-it-artemis-iq2-m
+- url: github:mudler/LocalAI/gallery/virtual.yaml@master
+  name: gemma-4-e2b-it-artemis-q3-k-m
+  description: |
+    TurinTech Artemis Q3_K_M quantization of Gemma 4 E2B from Google's QAT
+    checkpoint, for local chat and tool calls. Accepts image input through
+    the bundled F16 vision projector. Uses the embedded chat template and
+    an 8192-token context. Thinking is on by default.
+  license: apache-2.0
+  urls:
+    - https://huggingface.co/google/gemma-4-E2B-it-qat-q4_0-unquantized
+    - https://huggingface.co/turintech/gemma-4-E2B-it-Artemis-GGUF
+  icon: https://huggingface.co/turintech/gemma-4-E2B-it-Artemis-GGUF/resolve/13a1c266e43df6e184779f8bc446ca6c66dca152/assets/turintech-logo.png
+  tags:
+    - llm
+    - gguf
+    - cpu
+    - gpu
+    - gemma
+    - gemma4
+    - gemma-4
+    - multimodal
+    - vision
+  files:
+    - filename: llama-cpp/models/gemma-4-E2B-it-Artemis-GGUF/gemma-4-E2B-it-Artemis-Q3_K_M.gguf
+      sha256: 6bdb901ce3c468c626c5481076b9c56bc69a35fc3ad08c78b2b6cbede5cbd9da
+      uri: https://huggingface.co/turintech/gemma-4-E2B-it-Artemis-GGUF/resolve/13a1c266e43df6e184779f8bc446ca6c66dca152/gemma-4-E2B-it-Artemis-Q3_K_M.gguf
+    - filename: llama-cpp/mmproj/gemma-4-E2B-it-Artemis-GGUF/mmproj-F16.gguf
+      sha256: 35be86a206b3573bd149e59b0ea8d8c371dab6f407e1b3f4f73d6408ff6ce6bf
+      uri: https://huggingface.co/turintech/gemma-4-E2B-it-Artemis-GGUF/resolve/13a1c266e43df6e184779f8bc446ca6c66dca152/mmproj-F16.gguf
+  overrides:
+    backend: llama-cpp
+    context_size: 8192
+    function:
+      automatic_tool_parsing_fallback: true
+      grammar:
+        disable: true
+    known_usecases:
+      - chat
+      - vision
+    mmproj: llama-cpp/mmproj/gemma-4-E2B-it-Artemis-GGUF/mmproj-F16.gguf
+    options:
+      - use_jinja:true
+    parameters:
+      model: llama-cpp/models/gemma-4-E2B-it-Artemis-GGUF/gemma-4-E2B-it-Artemis-Q3_K_M.gguf
+    template:
+      use_tokenizer_template: true
+- url: github:mudler/LocalAI/gallery/virtual.yaml@master
+  name: gemma-4-e2b-it-artemis-iq3-xxs
+  description: |
+    TurinTech Artemis IQ3_XXS quantization of Gemma 4 E2B from Google's QAT
+    checkpoint, for local chat and tool calls. Accepts image input through
+    the bundled F16 vision projector. Uses the embedded chat template and
+    an 8192-token context. Thinking is on by default.
+  license: apache-2.0
+  urls:
+    - https://huggingface.co/google/gemma-4-E2B-it-qat-q4_0-unquantized
+    - https://huggingface.co/turintech/gemma-4-E2B-it-Artemis-GGUF
+  icon: https://huggingface.co/turintech/gemma-4-E2B-it-Artemis-GGUF/resolve/13a1c266e43df6e184779f8bc446ca6c66dca152/assets/turintech-logo.png
+  tags:
+    - llm
+    - gguf
+    - cpu
+    - gpu
+    - gemma
+    - gemma4
+    - gemma-4
+    - multimodal
+    - vision
+  files:
+    - filename: llama-cpp/models/gemma-4-E2B-it-Artemis-GGUF/gemma-4-E2B-it-Artemis-IQ3_XXS.gguf
+      sha256: 66b7eebe3316775b6e0194b3b4da2888c32830d3c7e3c749b1e861d121e97aa9
+      uri: https://huggingface.co/turintech/gemma-4-E2B-it-Artemis-GGUF/resolve/13a1c266e43df6e184779f8bc446ca6c66dca152/gemma-4-E2B-it-Artemis-IQ3_XXS.gguf
+    - filename: llama-cpp/mmproj/gemma-4-E2B-it-Artemis-GGUF/mmproj-F16.gguf
+      sha256: 35be86a206b3573bd149e59b0ea8d8c371dab6f407e1b3f4f73d6408ff6ce6bf
+      uri: https://huggingface.co/turintech/gemma-4-E2B-it-Artemis-GGUF/resolve/13a1c266e43df6e184779f8bc446ca6c66dca152/mmproj-F16.gguf
+  overrides:
+    backend: llama-cpp
+    context_size: 8192
+    function:
+      automatic_tool_parsing_fallback: true
+      grammar:
+        disable: true
+    known_usecases:
+      - chat
+      - vision
+    mmproj: llama-cpp/mmproj/gemma-4-E2B-it-Artemis-GGUF/mmproj-F16.gguf
+    options:
+      - use_jinja:true
+    parameters:
+      model: llama-cpp/models/gemma-4-E2B-it-Artemis-GGUF/gemma-4-E2B-it-Artemis-IQ3_XXS.gguf
+    template:
+      use_tokenizer_template: true
+- url: github:mudler/LocalAI/gallery/virtual.yaml@master
+  name: gemma-4-e2b-it-artemis-iq2-m
+  description: |
+    TurinTech Artemis IQ2_M quantization of Gemma 4 E2B from Google's QAT
+    checkpoint, for local chat and tool calls. Accepts image input through
+    the bundled F16 vision projector. Uses the embedded chat template and
+    an 8192-token context. Thinking is on by default.
+  license: apache-2.0
+  urls:
+    - https://huggingface.co/google/gemma-4-E2B-it-qat-q4_0-unquantized
+    - https://huggingface.co/turintech/gemma-4-E2B-it-Artemis-GGUF
+  icon: https://huggingface.co/turintech/gemma-4-E2B-it-Artemis-GGUF/resolve/13a1c266e43df6e184779f8bc446ca6c66dca152/assets/turintech-logo.png
+  tags:
+    - llm
+    - gguf
+    - cpu
+    - gpu
+    - gemma
+    - gemma4
+    - gemma-4
+    - multimodal
+    - vision
+  files:
+    - filename: llama-cpp/models/gemma-4-E2B-it-Artemis-GGUF/gemma-4-E2B-it-Artemis-IQ2_M.gguf
+      sha256: 61ecb5b0d2854afbc1436db7598465f77ccf63ec62210e75cc761e09ecb6d729
+      uri: https://huggingface.co/turintech/gemma-4-E2B-it-Artemis-GGUF/resolve/13a1c266e43df6e184779f8bc446ca6c66dca152/gemma-4-E2B-it-Artemis-IQ2_M.gguf
+    - filename: llama-cpp/mmproj/gemma-4-E2B-it-Artemis-GGUF/mmproj-F16.gguf
+      sha256: 35be86a206b3573bd149e59b0ea8d8c371dab6f407e1b3f4f73d6408ff6ce6bf
+      uri: https://huggingface.co/turintech/gemma-4-E2B-it-Artemis-GGUF/resolve/13a1c266e43df6e184779f8bc446ca6c66dca152/mmproj-F16.gguf
+  overrides:
+    backend: llama-cpp
+    context_size: 8192
+    function:
+      automatic_tool_parsing_fallback: true
+      grammar:
+        disable: true
+    known_usecases:
+      - chat
+      - vision
+    mmproj: llama-cpp/mmproj/gemma-4-E2B-it-Artemis-GGUF/mmproj-F16.gguf
+    options:
+      - use_jinja:true
+    parameters:
+      model: llama-cpp/models/gemma-4-E2B-it-Artemis-GGUF/gemma-4-E2B-it-Artemis-IQ2_M.gguf
+    template:
+      use_tokenizer_template: true
+- url: github:mudler/LocalAI/gallery/virtual.yaml@master
+  name: gemma-4-e4b-it-artemis-q4-0
+  description: |
+    TurinTech Artemis Q4_0 quantization of Gemma 4 E4B from Google's QAT
+    checkpoint, for local chat, reasoning, and tool use. Accepts image
+    input through the bundled F16 vision projector. Uses the embedded chat
+    template and an 8192-token context. Thinking is on by default.
+  license: apache-2.0
+  urls:
+    - https://huggingface.co/google/gemma-4-E4B-it-qat-q4_0-unquantized
+    - https://huggingface.co/turintech/gemma-4-E4B-it-Artemis-GGUF
+  icon: https://huggingface.co/turintech/gemma-4-E4B-it-Artemis-GGUF/resolve/31d9aa277c62a7036cde4e9aaa3c2d330e7775cf/assets/turintech-logo.png
+  tags:
+    - llm
+    - gguf
+    - cpu
+    - gpu
+    - gemma
+    - gemma4
+    - gemma-4
+    - multimodal
+    - vision
+  files:
+    - filename: llama-cpp/models/gemma-4-E4B-it-Artemis-GGUF/gemma-4-E4B-it-Artemis-Q4_0.gguf
+      sha256: 1d23b7340a5828d12c50ee60a9fd0662c6cf23117d4cd4f6a7e3ec1bfb503260
+      uri: https://huggingface.co/turintech/gemma-4-E4B-it-Artemis-GGUF/resolve/31d9aa277c62a7036cde4e9aaa3c2d330e7775cf/gemma-4-E4B-it-Artemis-Q4_0.gguf
+    - filename: llama-cpp/mmproj/gemma-4-E4B-it-Artemis-GGUF/mmproj-F16.gguf
+      sha256: b2abc3148c97f6d0ad820cef6b37db8cd1477eccb949f5722a8976c2f942c135
+      uri: https://huggingface.co/turintech/gemma-4-E4B-it-Artemis-GGUF/resolve/31d9aa277c62a7036cde4e9aaa3c2d330e7775cf/mmproj-F16.gguf
+  overrides:
+    backend: llama-cpp
+    context_size: 8192
+    function:
+      automatic_tool_parsing_fallback: true
+      grammar:
+        disable: true
+    known_usecases:
+      - chat
+      - vision
+    mmproj: llama-cpp/mmproj/gemma-4-E4B-it-Artemis-GGUF/mmproj-F16.gguf
+    options:
+      - use_jinja:true
+    parameters:
+      model: llama-cpp/models/gemma-4-E4B-it-Artemis-GGUF/gemma-4-E4B-it-Artemis-Q4_0.gguf
+    template:
+      use_tokenizer_template: true
+  variants:
+    - model: gemma-4-e4b-it-artemis-iq3-m
+    - model: gemma-4-e4b-it-artemis-q3-k-m
+- url: github:mudler/LocalAI/gallery/virtual.yaml@master
+  name: gemma-4-e4b-it-artemis-iq3-m
+  description: |
+    TurinTech Artemis IQ3_M quantization of Gemma 4 E4B from Google's QAT
+    checkpoint, for local chat, reasoning, and tool use. Accepts image
+    input through the bundled F16 vision projector. Uses the embedded chat
+    template and an 8192-token context. Thinking is on by default.
+  license: apache-2.0
+  urls:
+    - https://huggingface.co/google/gemma-4-E4B-it-qat-q4_0-unquantized
+    - https://huggingface.co/turintech/gemma-4-E4B-it-Artemis-GGUF
+  icon: https://huggingface.co/turintech/gemma-4-E4B-it-Artemis-GGUF/resolve/31d9aa277c62a7036cde4e9aaa3c2d330e7775cf/assets/turintech-logo.png
+  tags:
+    - llm
+    - gguf
+    - cpu
+    - gpu
+    - gemma
+    - gemma4
+    - gemma-4
+    - multimodal
+    - vision
+  files:
+    - filename: llama-cpp/models/gemma-4-E4B-it-Artemis-GGUF/gemma-4-E4B-it-Artemis-IQ3_M.gguf
+      sha256: 7d19112172d02bfe3edd711627bb988330e3fea8714c77362e2b77ae1634ac39
+      uri: https://huggingface.co/turintech/gemma-4-E4B-it-Artemis-GGUF/resolve/31d9aa277c62a7036cde4e9aaa3c2d330e7775cf/gemma-4-E4B-it-Artemis-IQ3_M.gguf
+    - filename: llama-cpp/mmproj/gemma-4-E4B-it-Artemis-GGUF/mmproj-F16.gguf
+      sha256: b2abc3148c97f6d0ad820cef6b37db8cd1477eccb949f5722a8976c2f942c135
+      uri: https://huggingface.co/turintech/gemma-4-E4B-it-Artemis-GGUF/resolve/31d9aa277c62a7036cde4e9aaa3c2d330e7775cf/mmproj-F16.gguf
+  overrides:
+    backend: llama-cpp
+    context_size: 8192
+    function:
+      automatic_tool_parsing_fallback: true
+      grammar:
+        disable: true
+    known_usecases:
+      - chat
+      - vision
+    mmproj: llama-cpp/mmproj/gemma-4-E4B-it-Artemis-GGUF/mmproj-F16.gguf
+    options:
+      - use_jinja:true
+    parameters:
+      model: llama-cpp/models/gemma-4-E4B-it-Artemis-GGUF/gemma-4-E4B-it-Artemis-IQ3_M.gguf
+    template:
+      use_tokenizer_template: true
+- url: github:mudler/LocalAI/gallery/virtual.yaml@master
+  name: gemma-4-e4b-it-artemis-q3-k-m
+  description: |
+    TurinTech Artemis Q3_K_M quantization of Gemma 4 E4B from Google's QAT
+    checkpoint, for local chat, reasoning, and tool use. Accepts image
+    input through the bundled F16 vision projector. Uses the embedded chat
+    template and an 8192-token context. Thinking is on by default.
+  license: apache-2.0
+  urls:
+    - https://huggingface.co/google/gemma-4-E4B-it-qat-q4_0-unquantized
+    - https://huggingface.co/turintech/gemma-4-E4B-it-Artemis-GGUF
+  icon: https://huggingface.co/turintech/gemma-4-E4B-it-Artemis-GGUF/resolve/31d9aa277c62a7036cde4e9aaa3c2d330e7775cf/assets/turintech-logo.png
+  tags:
+    - llm
+    - gguf
+    - cpu
+    - gpu
+    - gemma
+    - gemma4
+    - gemma-4
+    - multimodal
+    - vision
+  files:
+    - filename: llama-cpp/models/gemma-4-E4B-it-Artemis-GGUF/gemma-4-E4B-it-Artemis-Q3_K_M.gguf
+      sha256: 182f4a4f29461620f7404b9eb25a03bd7fe9f91de49718662c9146c01c9d3547
+      uri: https://huggingface.co/turintech/gemma-4-E4B-it-Artemis-GGUF/resolve/31d9aa277c62a7036cde4e9aaa3c2d330e7775cf/gemma-4-E4B-it-Artemis-Q3_K_M.gguf
+    - filename: llama-cpp/mmproj/gemma-4-E4B-it-Artemis-GGUF/mmproj-F16.gguf
+      sha256: b2abc3148c97f6d0ad820cef6b37db8cd1477eccb949f5722a8976c2f942c135
+      uri: https://huggingface.co/turintech/gemma-4-E4B-it-Artemis-GGUF/resolve/31d9aa277c62a7036cde4e9aaa3c2d330e7775cf/mmproj-F16.gguf
+  overrides:
+    backend: llama-cpp
+    context_size: 8192
+    function:
+      automatic_tool_parsing_fallback: true
+      grammar:
+        disable: true
+    known_usecases:
+      - chat
+      - vision
+    mmproj: llama-cpp/mmproj/gemma-4-E4B-it-Artemis-GGUF/mmproj-F16.gguf
+    options:
+      - use_jinja:true
+    parameters:
+      model: llama-cpp/models/gemma-4-E4B-it-Artemis-GGUF/gemma-4-E4B-it-Artemis-Q3_K_M.gguf
+    template:
+      use_tokenizer_template: true
 - url: github:mudler/LocalAI/gallery/virtual.yaml@master
   name: qwen3.5-0.8b-artemis-q5
   description: |
diff --git a/gallery/models/index.yaml b/gallery/models/index.yaml
index ea5f22a7a7..5ef216d461 100644
--- a/gallery/models/index.yaml
+++ b/gallery/models/index.yaml
@@ -1667,6 +1667,391 @@
         - pooling:cls
       parameters:
         model: bge-small-en-v1.5-q4_k_m.gguf
+- kind: model
+  id: gemma-4-e2b-it-artemis-q3-k-l
+  name: gemma-4-e2b-it-artemis-q3-k-l
+  version: 0.0.0-gallery.a49a9ec13280
+  description: |
+    TurinTech Artemis Q3_K_L quantization of Gemma 4 E2B from Google's QAT
+    checkpoint, for local chat and tool calls. Accepts image input through
+    the bundled F16 vision projector. Uses the embedded chat template and
+    an 8192-token context. Thinking is on by default.
+  license: apache-2.0
+  urls:
+    - https://huggingface.co/google/gemma-4-E2B-it-qat-q4_0-unquantized
+    - https://huggingface.co/turintech/gemma-4-E2B-it-Artemis-GGUF
+  tags:
+    - llm
+    - gguf
+    - cpu
+    - gpu
+    - gemma
+    - gemma4
+    - gemma-4
+    - multimodal
+    - vision
+  icon: https://huggingface.co/turintech/gemma-4-E2B-it-Artemis-GGUF/resolve/13a1c266e43df6e184779f8bc446ca6c66dca152/assets/turintech-logo.png
+  channel: gallery
+  policy:
+    requireSignature: false
+  schemaVersion: 1
+  model:
+    url: github:mudler/LocalAI/gallery/virtual.yaml@master
+    files:
+      - filename: llama-cpp/models/gemma-4-E2B-it-Artemis-GGUF/gemma-4-E2B-it-Artemis-Q3_K_L.gguf
+        sha256: 1e207a449e37ea902de237aba1158f960a91bd344bee6edc5b0aa0dbce015621
+        uri: https://huggingface.co/turintech/gemma-4-E2B-it-Artemis-GGUF/resolve/13a1c266e43df6e184779f8bc446ca6c66dca152/gemma-4-E2B-it-Artemis-Q3_K_L.gguf
+      - filename: llama-cpp/mmproj/gemma-4-E2B-it-Artemis-GGUF/mmproj-F16.gguf
+        sha256: 35be86a206b3573bd149e59b0ea8d8c371dab6f407e1b3f4f73d6408ff6ce6bf
+        uri: https://huggingface.co/turintech/gemma-4-E2B-it-Artemis-GGUF/resolve/13a1c266e43df6e184779f8bc446ca6c66dca152/mmproj-F16.gguf
+    overrides:
+      backend: llama-cpp
+      context_size: 8192
+      function:
+        automatic_tool_parsing_fallback: true
+        grammar:
+          disable: true
+      known_usecases:
+        - chat
+        - vision
+      mmproj: llama-cpp/mmproj/gemma-4-E2B-it-Artemis-GGUF/mmproj-F16.gguf
+      options:
+        - use_jinja:true
+      parameters:
+        model: llama-cpp/models/gemma-4-E2B-it-Artemis-GGUF/gemma-4-E2B-it-Artemis-Q3_K_L.gguf
+      template:
+        use_tokenizer_template: true
+    variants:
+      - model: gemma-4-e2b-it-artemis-q3-k-m
+      - model: gemma-4-e2b-it-artemis-iq3-xxs
+      - model: gemma-4-e2b-it-artemis-iq2-m
+- kind: model
+  id: gemma-4-e2b-it-artemis-q3-k-m
+  name: gemma-4-e2b-it-artemis-q3-k-m
+  version: 0.0.0-gallery.7d34154c17da
+  description: |
+    TurinTech Artemis Q3_K_M quantization of Gemma 4 E2B from Google's QAT
+    checkpoint, for local chat and tool calls. Accepts image input through
+    the bundled F16 vision projector. Uses the embedded chat template and
+    an 8192-token context. Thinking is on by default.
+  license: apache-2.0
+  urls:
+    - https://huggingface.co/google/gemma-4-E2B-it-qat-q4_0-unquantized
+    - https://huggingface.co/turintech/gemma-4-E2B-it-Artemis-GGUF
+  tags:
+    - llm
+    - gguf
+    - cpu
+    - gpu
+    - gemma
+    - gemma4
+    - gemma-4
+    - multimodal
+    - vision
+  icon: https://huggingface.co/turintech/gemma-4-E2B-it-Artemis-GGUF/resolve/13a1c266e43df6e184779f8bc446ca6c66dca152/assets/turintech-logo.png
+  channel: gallery
+  policy:
+    requireSignature: false
+  schemaVersion: 1
+  model:
+    url: github:mudler/LocalAI/gallery/virtual.yaml@master
+    files:
+      - filename: llama-cpp/models/gemma-4-E2B-it-Artemis-GGUF/gemma-4-E2B-it-Artemis-Q3_K_M.gguf
+        sha256: 6bdb901ce3c468c626c5481076b9c56bc69a35fc3ad08c78b2b6cbede5cbd9da
+        uri: https://huggingface.co/turintech/gemma-4-E2B-it-Artemis-GGUF/resolve/13a1c266e43df6e184779f8bc446ca6c66dca152/gemma-4-E2B-it-Artemis-Q3_K_M.gguf
+      - filename: llama-cpp/mmproj/gemma-4-E2B-it-Artemis-GGUF/mmproj-F16.gguf
+        sha256: 35be86a206b3573bd149e59b0ea8d8c371dab6f407e1b3f4f73d6408ff6ce6bf
+        uri: https://huggingface.co/turintech/gemma-4-E2B-it-Artemis-GGUF/resolve/13a1c266e43df6e184779f8bc446ca6c66dca152/mmproj-F16.gguf
+    overrides:
+      backend: llama-cpp
+      context_size: 8192
+      function:
+        automatic_tool_parsing_fallback: true
+        grammar:
+          disable: true
+      known_usecases:
+        - chat
+        - vision
+      mmproj: llama-cpp/mmproj/gemma-4-E2B-it-Artemis-GGUF/mmproj-F16.gguf
+      options:
+        - use_jinja:true
+      parameters:
+        model: llama-cpp/models/gemma-4-E2B-it-Artemis-GGUF/gemma-4-E2B-it-Artemis-Q3_K_M.gguf
+      template:
+        use_tokenizer_template: true
+- kind: model
+  id: gemma-4-e2b-it-artemis-iq3-xxs
+  name: gemma-4-e2b-it-artemis-iq3-xxs
+  version: 0.0.0-gallery.c477c155a918
+  description: |
+    TurinTech Artemis IQ3_XXS quantization of Gemma 4 E2B from Google's QAT
+    checkpoint, for local chat and tool calls. Accepts image input through
+    the bundled F16 vision projector. Uses the embedded chat template and
+    an 8192-token context. Thinking is on by default.
+  license: apache-2.0
+  urls:
+    - https://huggingface.co/google/gemma-4-E2B-it-qat-q4_0-unquantized
+    - https://huggingface.co/turintech/gemma-4-E2B-it-Artemis-GGUF
+  tags:
+    - llm
+    - gguf
+    - cpu
+    - gpu
+    - gemma
+    - gemma4
+    - gemma-4
+    - multimodal
+    - vision
+  icon: https://huggingface.co/turintech/gemma-4-E2B-it-Artemis-GGUF/resolve/13a1c266e43df6e184779f8bc446ca6c66dca152/assets/turintech-logo.png
+  channel: gallery
+  policy:
+    requireSignature: false
+  schemaVersion: 1
+  model:
+    url: github:mudler/LocalAI/gallery/virtual.yaml@master
+    files:
+      - filename: llama-cpp/models/gemma-4-E2B-it-Artemis-GGUF/gemma-4-E2B-it-Artemis-IQ3_XXS.gguf
+        sha256: 66b7eebe3316775b6e0194b3b4da2888c32830d3c7e3c749b1e861d121e97aa9
+        uri: https://huggingface.co/turintech/gemma-4-E2B-it-Artemis-GGUF/resolve/13a1c266e43df6e184779f8bc446ca6c66dca152/gemma-4-E2B-it-Artemis-IQ3_XXS.gguf
+      - filename: llama-cpp/mmproj/gemma-4-E2B-it-Artemis-GGUF/mmproj-F16.gguf
+        sha256: 35be86a206b3573bd149e59b0ea8d8c371dab6f407e1b3f4f73d6408ff6ce6bf
+        uri: https://huggingface.co/turintech/gemma-4-E2B-it-Artemis-GGUF/resolve/13a1c266e43df6e184779f8bc446ca6c66dca152/mmproj-F16.gguf
+    overrides:
+      backend: llama-cpp
+      context_size: 8192
+      function:
+        automatic_tool_parsing_fallback: true
+        grammar:
+          disable: true
+      known_usecases:
+        - chat
+        - vision
+      mmproj: llama-cpp/mmproj/gemma-4-E2B-it-Artemis-GGUF/mmproj-F16.gguf
+      options:
+        - use_jinja:true
+      parameters:
+        model: llama-cpp/models/gemma-4-E2B-it-Artemis-GGUF/gemma-4-E2B-it-Artemis-IQ3_XXS.gguf
+      template:
+        use_tokenizer_template: true
+- kind: model
+  id: gemma-4-e2b-it-artemis-iq2-m
+  name: gemma-4-e2b-it-artemis-iq2-m
+  version: 0.0.0-gallery.6ad57d354aa3
+  description: |
+    TurinTech Artemis IQ2_M quantization of Gemma 4 E2B from Google's QAT
+    checkpoint, for local chat and tool calls. Accepts image input through
+    the bundled F16 vision projector. Uses the embedded chat template and
+    an 8192-token context. Thinking is on by default.
+  license: apache-2.0
+  urls:
+    - https://huggingface.co/google/gemma-4-E2B-it-qat-q4_0-unquantized
+    - https://huggingface.co/turintech/gemma-4-E2B-it-Artemis-GGUF
+  tags:
+    - llm
+    - gguf
+    - cpu
+    - gpu
+    - gemma
+    - gemma4
+    - gemma-4
+    - multimodal
+    - vision
+  icon: https://huggingface.co/turintech/gemma-4-E2B-it-Artemis-GGUF/resolve/13a1c266e43df6e184779f8bc446ca6c66dca152/assets/turintech-logo.png
+  channel: gallery
+  policy:
+    requireSignature: false
+  schemaVersion: 1
+  model:
+    url: github:mudler/LocalAI/gallery/virtual.yaml@master
+    files:
+      - filename: llama-cpp/models/gemma-4-E2B-it-Artemis-GGUF/gemma-4-E2B-it-Artemis-IQ2_M.gguf
+        sha256: 61ecb5b0d2854afbc1436db7598465f77ccf63ec62210e75cc761e09ecb6d729
+        uri: https://huggingface.co/turintech/gemma-4-E2B-it-Artemis-GGUF/resolve/13a1c266e43df6e184779f8bc446ca6c66dca152/gemma-4-E2B-it-Artemis-IQ2_M.gguf
+      - filename: llama-cpp/mmproj/gemma-4-E2B-it-Artemis-GGUF/mmproj-F16.gguf
+        sha256: 35be86a206b3573bd149e59b0ea8d8c371dab6f407e1b3f4f73d6408ff6ce6bf
+        uri: https://huggingface.co/turintech/gemma-4-E2B-it-Artemis-GGUF/resolve/13a1c266e43df6e184779f8bc446ca6c66dca152/mmproj-F16.gguf
+    overrides:
+      backend: llama-cpp
+      context_size: 8192
+      function:
+        automatic_tool_parsing_fallback: true
+        grammar:
+          disable: true
+      known_usecases:
+        - chat
+        - vision
+      mmproj: llama-cpp/mmproj/gemma-4-E2B-it-Artemis-GGUF/mmproj-F16.gguf
+      options:
+        - use_jinja:true
+      parameters:
+        model: llama-cpp/models/gemma-4-E2B-it-Artemis-GGUF/gemma-4-E2B-it-Artemis-IQ2_M.gguf
+      template:
+        use_tokenizer_template: true
+- kind: model
+  id: gemma-4-e4b-it-artemis-q4-0
+  name: gemma-4-e4b-it-artemis-q4-0
+  version: 0.0.0-gallery.f840da8f3f6e
+  description: |
+    TurinTech Artemis Q4_0 quantization of Gemma 4 E4B from Google's QAT
+    checkpoint, for local chat, reasoning, and tool use. Accepts image
+    input through the bundled F16 vision projector. Uses the embedded chat
+    template and an 8192-token context. Thinking is on by default.
+  license: apache-2.0
+  urls:
+    - https://huggingface.co/google/gemma-4-E4B-it-qat-q4_0-unquantized
+    - https://huggingface.co/turintech/gemma-4-E4B-it-Artemis-GGUF
+  tags:
+    - llm
+    - gguf
+    - cpu
+    - gpu
+    - gemma
+    - gemma4
+    - gemma-4
+    - multimodal
+    - vision
+  icon: https://huggingface.co/turintech/gemma-4-E4B-it-Artemis-GGUF/resolve/31d9aa277c62a7036cde4e9aaa3c2d330e7775cf/assets/turintech-logo.png
+  channel: gallery
+  policy:
+    requireSignature: false
+  schemaVersion: 1
+  model:
+    url: github:mudler/LocalAI/gallery/virtual.yaml@master
+    files:
+      - filename: llama-cpp/models/gemma-4-E4B-it-Artemis-GGUF/gemma-4-E4B-it-Artemis-Q4_0.gguf
+        sha256: 1d23b7340a5828d12c50ee60a9fd0662c6cf23117d4cd4f6a7e3ec1bfb503260
+        uri: https://huggingface.co/turintech/gemma-4-E4B-it-Artemis-GGUF/resolve/31d9aa277c62a7036cde4e9aaa3c2d330e7775cf/gemma-4-E4B-it-Artemis-Q4_0.gguf
+      - filename: llama-cpp/mmproj/gemma-4-E4B-it-Artemis-GGUF/mmproj-F16.gguf
+        sha256: b2abc3148c97f6d0ad820cef6b37db8cd1477eccb949f5722a8976c2f942c135
+        uri: https://huggingface.co/turintech/gemma-4-E4B-it-Artemis-GGUF/resolve/31d9aa277c62a7036cde4e9aaa3c2d330e7775cf/mmproj-F16.gguf
+    overrides:
+      backend: llama-cpp
+      context_size: 8192
+      function:
+        automatic_tool_parsing_fallback: true
+        grammar:
+          disable: true
+      known_usecases:
+        - chat
+        - vision
+      mmproj: llama-cpp/mmproj/gemma-4-E4B-it-Artemis-GGUF/mmproj-F16.gguf
+      options:
+        - use_jinja:true
+      parameters:
+        model: llama-cpp/models/gemma-4-E4B-it-Artemis-GGUF/gemma-4-E4B-it-Artemis-Q4_0.gguf
+      template:
+        use_tokenizer_template: true
+    variants:
+      - model: gemma-4-e4b-it-artemis-iq3-m
+      - model: gemma-4-e4b-it-artemis-q3-k-m
+- kind: model
+  id: gemma-4-e4b-it-artemis-iq3-m
+  name: gemma-4-e4b-it-artemis-iq3-m
+  version: 0.0.0-gallery.f622a887acdd
+  description: |
+    TurinTech Artemis IQ3_M quantization of Gemma 4 E4B from Google's QAT
+    checkpoint, for local chat, reasoning, and tool use. Accepts image
+    input through the bundled F16 vision projector. Uses the embedded chat
+    template and an 8192-token context. Thinking is on by default.
+  license: apache-2.0
+  urls:
+    - https://huggingface.co/google/gemma-4-E4B-it-qat-q4_0-unquantized
+    - https://huggingface.co/turintech/gemma-4-E4B-it-Artemis-GGUF
+  tags:
+    - llm
+    - gguf
+    - cpu
+    - gpu
+    - gemma
+    - gemma4
+    - gemma-4
+    - multimodal
+    - vision
+  icon: https://huggingface.co/turintech/gemma-4-E4B-it-Artemis-GGUF/resolve/31d9aa277c62a7036cde4e9aaa3c2d330e7775cf/assets/turintech-logo.png
+  channel: gallery
+  policy:
+    requireSignature: false
+  schemaVersion: 1
+  model:
+    url: github:mudler/LocalAI/gallery/virtual.yaml@master
+    files:
+      - filename: llama-cpp/models/gemma-4-E4B-it-Artemis-GGUF/gemma-4-E4B-it-Artemis-IQ3_M.gguf
+        sha256: 7d19112172d02bfe3edd711627bb988330e3fea8714c77362e2b77ae1634ac39
+        uri: https://huggingface.co/turintech/gemma-4-E4B-it-Artemis-GGUF/resolve/31d9aa277c62a7036cde4e9aaa3c2d330e7775cf/gemma-4-E4B-it-Artemis-IQ3_M.gguf
+      - filename: llama-cpp/mmproj/gemma-4-E4B-it-Artemis-GGUF/mmproj-F16.gguf
+        sha256: b2abc3148c97f6d0ad820cef6b37db8cd1477eccb949f5722a8976c2f942c135
+        uri: https://huggingface.co/turintech/gemma-4-E4B-it-Artemis-GGUF/resolve/31d9aa277c62a7036cde4e9aaa3c2d330e7775cf/mmproj-F16.gguf
+    overrides:
+      backend: llama-cpp
+      context_size: 8192
+      function:
+        automatic_tool_parsing_fallback: true
+        grammar:
+          disable: true
+      known_usecases:
+        - chat
+        - vision
+      mmproj: llama-cpp/mmproj/gemma-4-E4B-it-Artemis-GGUF/mmproj-F16.gguf
+      options:
+        - use_jinja:true
+      parameters:
+        model: llama-cpp/models/gemma-4-E4B-it-Artemis-GGUF/gemma-4-E4B-it-Artemis-IQ3_M.gguf
+      template:
+        use_tokenizer_template: true
+- kind: model
+  id: gemma-4-e4b-it-artemis-q3-k-m
+  name: gemma-4-e4b-it-artemis-q3-k-m
+  version: 0.0.0-gallery.9c7e3af29345
+  description: |
+    TurinTech Artemis Q3_K_M quantization of Gemma 4 E4B from Google's QAT
+    checkpoint, for local chat, reasoning, and tool use. Accepts image
+    input through the bundled F16 vision projector. Uses the embedded chat
+    template and an 8192-token context. Thinking is on by default.
+  license: apache-2.0
+  urls:
+    - https://huggingface.co/google/gemma-4-E4B-it-qat-q4_0-unquantized
+    - https://huggingface.co/turintech/gemma-4-E4B-it-Artemis-GGUF
+  tags:
+    - llm
+    - gguf
+    - cpu
+    - gpu
+    - gemma
+    - gemma4
+    - gemma-4
+    - multimodal
+    - vision
+  icon: https://huggingface.co/turintech/gemma-4-E4B-it-Artemis-GGUF/resolve/31d9aa277c62a7036cde4e9aaa3c2d330e7775cf/assets/turintech-logo.png
+  channel: gallery
+  policy:
+    requireSignature: false
+  schemaVersion: 1
+  model:
+    url: github:mudler/LocalAI/gallery/virtual.yaml@master
+    files:
+      - filename: llama-cpp/models/gemma-4-E4B-it-Artemis-GGUF/gemma-4-E4B-it-Artemis-Q3_K_M.gguf
+        sha256: 182f4a4f29461620f7404b9eb25a03bd7fe9f91de49718662c9146c01c9d3547
+        uri: https://huggingface.co/turintech/gemma-4-E4B-it-Artemis-GGUF/resolve/31d9aa277c62a7036cde4e9aaa3c2d330e7775cf/gemma-4-E4B-it-Artemis-Q3_K_M.gguf
+      - filename: llama-cpp/mmproj/gemma-4-E4B-it-Artemis-GGUF/mmproj-F16.gguf
+        sha256: b2abc3148c97f6d0ad820cef6b37db8cd1477eccb949f5722a8976c2f942c135
+        uri: https://huggingface.co/turintech/gemma-4-E4B-it-Artemis-GGUF/resolve/31d9aa277c62a7036cde4e9aaa3c2d330e7775cf/mmproj-F16.gguf
+    overrides:
+      backend: llama-cpp
+      context_size: 8192
+      function:
+        automatic_tool_parsing_fallback: true
+        grammar:
+          disable: true
+      known_usecases:
+        - chat
+        - vision
+      mmproj: llama-cpp/mmproj/gemma-4-E4B-it-Artemis-GGUF/mmproj-F16.gguf
+      options:
+        - use_jinja:true
+      parameters:
+        model: llama-cpp/models/gemma-4-E4B-it-Artemis-GGUF/gemma-4-E4B-it-Artemis-Q3_K_M.gguf
+      template:
+        use_tokenizer_template: true
 - kind: model
   id: qwen3.5-0.8b-artemis-q5
   name: qwen3.5-0.8b-artemis-q5