chore(model gallery): add voxtral (which is only available in development) (#8532)

Signed-off-by: Ettore Di Giacinto <mudler@localai.io>

gallery/index.yaml

@@ -1,4 +1,31 @@
 ---
+- url: "github:mudler/LocalAI/gallery/voxtral.yaml@master"
+  name: voxtral-mini-4b-realtime
+  license: apache-2.0
+  description: |
+    Voxtral Mini 4B Realtime is a speech-to-text model from Mistral AI. It is a 4B parameter model optimized for fast, accurate audio transcription with low latency, making it ideal for real-time applications. The model uses the Voxtral architecture for efficient audio processing.
+  urls:
+    - https://huggingface.co/mistralai/Voxtral-Mini-4B-Realtime-2602
+    - https://github.com/antirez/voxtral.c
+  tags:
+    - stt
+    - speech-to-text
+    - audio-transcription
+    - cpu
+    - metal
+    - mistral
+  overrides:
+    parameters:
+      model: voxtral-model
+  files:
+    - filename: voxtral-model/consolidated.safetensors
+      uri: https://huggingface.co/mistralai/Voxtral-Mini-4B-Realtime-2602/resolve/main/consolidated.safetensors
+      sha256: 263f178fe752c90a2ae58f037a95ed092db8b14768b0978b8c48f66979c8345d
+    - filename: voxtral-model/params.json
+      uri: https://huggingface.co/mistralai/Voxtral-Mini-4B-Realtime-2602/resolve/main/params.json
+    - filename: voxtral-model/tekken.json
+      uri: https://huggingface.co/mistralai/Voxtral-Mini-4B-Realtime-2602/resolve/main/tekken.json
+      sha256: 8434af1d39eba99f0ef46cf1450bf1a63fa941a26933a1ef5dbbf4adf0d00e44
 - name: "ace-step-turbo"
   license: mit
   tags: