feat(gallery): add Spark-X2.6 4B variants

docs/content/features/model-gallery.md

@@ -272,6 +272,20 @@ These builds support text chat only: the publisher has no verified vision projec
 They use standard GGUF files without MTP decoding.
 See the [model card](https://huggingface.co/ukisai/Swift-1.5-Qwen3.8-27B-GSQ-RCO-GGUF) and [Swift Open License v1.0](https://huggingface.co/ukisai/Swift-1.5-Qwen3.8-27B-GSQ-RCO-GGUF/blob/main/LICENSE) for usage terms.
 
+## Spark-X2.6-4B
+
+Install Spark-X2.6-4B for text chat, reasoning, and coding with llama.cpp.
+The gallery groups the publisher's Q4_K_M and Q8_0 GGUF builds as variants:
+
+```bash
+local-ai models install spark-x2.6-4b-q4
+local-ai models install spark-x2.6-4b-q4 --variant spark-x2.6-4b-q8
+```
+
+Both builds use the embedded chat template and a 32,768-token default context.
+They use the same Spark architecture as the existing Spark-X2.5 models.
+See the [publisher's model card](https://huggingface.co/XHToken/Spark-X2.6-4B-GGUF) for model details and the Apache 2.0 license.
+
 ## Sharp-Spark-X2.5-4B
 
 Install `sharp-spark-x2.5-4b` for coding and text chat with llama.cpp.

gallery/index.yaml

@@ -3,6 +3,84 @@
 # 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: spark-x2.6-4b-q4
+  description: |
+    Spark-X2.6-4B is XHToken's 4B text model for conversation, reasoning,
+    coding, and multilingual tasks. This build uses Q4_K_M GGUF weights,
+    the embedded Jinja chat template, and a 32K-token default context.
+  license: apache-2.0
+  urls:
+    - https://huggingface.co/XHToken/Spark-X2.6-4B
+    - https://huggingface.co/XHToken/Spark-X2.6-4B-GGUF
+  tags:
+    - llm
+    - gguf
+    - cpu
+    - gpu
+    - reasoning
+    - coding
+    - multilingual
+  files:
+    - filename: llama-cpp/models/spark-x2.6-4b/Spark-X2.6-4B-Q4_K_M.gguf
+      sha256: 91cb2811aebb47ac916aa619d27d4cdd7f4143106b7917accb8ead4cd7f5cbd9
+      uri: https://huggingface.co/XHToken/Spark-X2.6-4B-GGUF/resolve/fe31f8d64890302cc3a1d365d700357b4b75ba77/Spark-X2.6-4B-Q4_K_M.gguf
+  overrides:
+    backend: llama-cpp
+    context_size: 32768
+    known_usecases:
+      - chat
+    options:
+      - use_jinja:true
+    parameters:
+      min_p: 0
+      model: llama-cpp/models/spark-x2.6-4b/Spark-X2.6-4B-Q4_K_M.gguf
+      repeat_penalty: 1
+      temperature: 1
+      top_k: -1
+      top_p: 0.95
+    template:
+      use_tokenizer_template: true
+  variants:
+    - model: spark-x2.6-4b-q8
+- url: github:mudler/LocalAI/gallery/virtual.yaml@master
+  name: spark-x2.6-4b-q8
+  description: |
+    Spark-X2.6-4B is XHToken's 4B text model for conversation, reasoning,
+    coding, and multilingual tasks. This build uses Q8_0 GGUF weights,
+    the embedded Jinja chat template, and a 32K-token default context.
+  license: apache-2.0
+  urls:
+    - https://huggingface.co/XHToken/Spark-X2.6-4B
+    - https://huggingface.co/XHToken/Spark-X2.6-4B-GGUF
+  tags:
+    - llm
+    - gguf
+    - cpu
+    - gpu
+    - reasoning
+    - coding
+    - multilingual
+  files:
+    - filename: llama-cpp/models/spark-x2.6-4b/Spark-X2.6-4B-Q8_0.gguf
+      sha256: 145c2aa606820663f26ad530b8fc141ed885e2e9ca29f9be7c3bfe8931d58b81
+      uri: https://huggingface.co/XHToken/Spark-X2.6-4B-GGUF/resolve/fe31f8d64890302cc3a1d365d700357b4b75ba77/Spark-X2.6-4B-Q8_0.gguf
+  overrides:
+    backend: llama-cpp
+    context_size: 32768
+    known_usecases:
+      - chat
+    options:
+      - use_jinja:true
+    parameters:
+      min_p: 0
+      model: llama-cpp/models/spark-x2.6-4b/Spark-X2.6-4B-Q8_0.gguf
+      repeat_penalty: 1
+      temperature: 1
+      top_k: -1
+      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 +4,100 @@
 # 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: spark-x2.6-4b-q4
+  name: spark-x2.6-4b-q4
+  version: 0.0.0-gallery.893c50e4aba3
+  description: |
+    Spark-X2.6-4B is XHToken's 4B text model for conversation, reasoning,
+    coding, and multilingual tasks. This build uses Q4_K_M GGUF weights,
+    the embedded Jinja chat template, and a 32K-token default context.
+  license: apache-2.0
+  urls:
+    - https://huggingface.co/XHToken/Spark-X2.6-4B
+    - https://huggingface.co/XHToken/Spark-X2.6-4B-GGUF
+  tags:
+    - llm
+    - gguf
+    - cpu
+    - gpu
+    - reasoning
+    - coding
+    - multilingual
+  channel: gallery
+  policy:
+    requireSignature: false
+  schemaVersion: 1
+  model:
+    url: github:mudler/LocalAI/gallery/virtual.yaml@master
+    files:
+      - filename: llama-cpp/models/spark-x2.6-4b/Spark-X2.6-4B-Q4_K_M.gguf
+        sha256: 91cb2811aebb47ac916aa619d27d4cdd7f4143106b7917accb8ead4cd7f5cbd9
+        uri: https://huggingface.co/XHToken/Spark-X2.6-4B-GGUF/resolve/fe31f8d64890302cc3a1d365d700357b4b75ba77/Spark-X2.6-4B-Q4_K_M.gguf
+    overrides:
+      backend: llama-cpp
+      context_size: 32768
+      known_usecases:
+        - chat
+      options:
+        - use_jinja:true
+      parameters:
+        min_p: 0
+        model: llama-cpp/models/spark-x2.6-4b/Spark-X2.6-4B-Q4_K_M.gguf
+        repeat_penalty: 1
+        temperature: 1
+        top_k: -1
+        top_p: 0.95
+      template:
+        use_tokenizer_template: true
+    variants:
+      - model: spark-x2.6-4b-q8
+- kind: model
+  id: spark-x2.6-4b-q8
+  name: spark-x2.6-4b-q8
+  version: 0.0.0-gallery.dc29344bd04b
+  description: |
+    Spark-X2.6-4B is XHToken's 4B text model for conversation, reasoning,
+    coding, and multilingual tasks. This build uses Q8_0 GGUF weights,
+    the embedded Jinja chat template, and a 32K-token default context.
+  license: apache-2.0
+  urls:
+    - https://huggingface.co/XHToken/Spark-X2.6-4B
+    - https://huggingface.co/XHToken/Spark-X2.6-4B-GGUF
+  tags:
+    - llm
+    - gguf
+    - cpu
+    - gpu
+    - reasoning
+    - coding
+    - multilingual
+  channel: gallery
+  policy:
+    requireSignature: false
+  schemaVersion: 1
+  model:
+    url: github:mudler/LocalAI/gallery/virtual.yaml@master
+    files:
+      - filename: llama-cpp/models/spark-x2.6-4b/Spark-X2.6-4B-Q8_0.gguf
+        sha256: 145c2aa606820663f26ad530b8fc141ed885e2e9ca29f9be7c3bfe8931d58b81
+        uri: https://huggingface.co/XHToken/Spark-X2.6-4B-GGUF/resolve/fe31f8d64890302cc3a1d365d700357b4b75ba77/Spark-X2.6-4B-Q8_0.gguf
+    overrides:
+      backend: llama-cpp
+      context_size: 32768
+      known_usecases:
+        - chat
+      options:
+        - use_jinja:true
+      parameters:
+        min_p: 0
+        model: llama-cpp/models/spark-x2.6-4b/Spark-X2.6-4B-Q8_0.gguf
+        repeat_penalty: 1
+        temperature: 1
+        top_k: -1
+        top_p: 0.95
+      template:
+        use_tokenizer_template: true
 - kind: model
   id: adavar-3b-q4
   name: adavar-3b-q4