chore(model gallery): 🤖 add 1 new models via gallery agent

gallery/index.yaml

@@ -3295,7 +3295,7 @@
     - gemma3
     - gemma-3
   overrides:
     #mmproj: gemma-3-27b-it-mmproj-f16.gguf
     parameters:
       model: gemma-3-27b-it-Q4_K_M.gguf
   files:
@@ -3313,7 +3313,7 @@
   description: |
     google/gemma-3-12b-it is an open-source, state-of-the-art, lightweight, multimodal model built from the same research and technology used to create the Gemini models. It is capable of handling text and image input and generating text output. It has a large context window of 128K tokens and supports over 140 languages. The 12B variant has been fine-tuned using the instruction-tuning approach. Gemma 3 models are suitable for a variety of text generation and image understanding tasks, including question answering, summarization, and reasoning. Their relatively small size makes them deployable in environments with limited resources such as laptops, desktops, or your own cloud infrastructure.
   overrides:
     #mmproj: gemma-3-12b-it-mmproj-f16.gguf
     parameters:
       model: gemma-3-12b-it-Q4_K_M.gguf
   files:
@@ -3331,7 +3331,7 @@
   description: |
     Gemma is a family of lightweight, state-of-the-art open models from Google, built from the same research and technology used to create the Gemini models. Gemma 3 models are multimodal, handling text and image input and generating text output, with open weights for both pre-trained variants and instruction-tuned variants. Gemma 3 has a large, 128K context window, multilingual support in over 140 languages, and is available in more sizes than previous versions. Gemma 3 models are well-suited for a variety of text generation and image understanding tasks, including question answering, summarization, and reasoning. Their relatively small size makes it possible to deploy them in environments with limited resources such as laptops, desktops or your own cloud infrastructure, democratizing access to state of the art AI models and helping foster innovation for everyone. Gemma-3-4b-it is a 4 billion parameter model.
   overrides:
     #mmproj: gemma-3-4b-it-mmproj-f16.gguf
     parameters:
       model: gemma-3-4b-it-Q4_K_M.gguf
   files:
@@ -6949,7 +6949,7 @@
       sha256: 2756551de7d8ff7093c2c5eec1cd00f1868bc128433af53f5a8d434091d4eb5a
       uri: huggingface://Triangle104/Nano_Imp_1B-Q8_0-GGUF/nano_imp_1b-q8_0.gguf
 - &qwen25
   name: "qwen2.5-14b-instruct" ## Qwen2.5
   icon: https://avatars.githubusercontent.com/u/141221163
   url: "github:mudler/LocalAI/gallery/chatml.yaml@master"
   license: apache-2.0
@@ -23049,3 +23049,45 @@
     - filename: YanoljaNEXT-Rosetta-27B-2511.i1-Q4_K_M.gguf
       sha256: 0a599099e93ad521045e17d82365a73c1738fff0603d6cb2c9557e96fbc907cb
       uri: huggingface://mradermacher/YanoljaNEXT-Rosetta-27B-2511-i1-GGUF/YanoljaNEXT-Rosetta-27B-2511.i1-Q4_K_M.gguf
+- !!merge <<: *qwen3vl
+  name: "qwen3-vl-12b-instruct-brainstorm20x"
+  urls:
+    - https://huggingface.co/mradermacher/Qwen3-VL-12B-Instruct-Brainstorm20x-GGUF
+  description: |
+    **Model Name:** Qwen3-VL-12B-Instruct-Brainstorm20x
+    **Base Model:** Qwen/Qwen3-VL-8B-Instruct
+    **Repository:** [mradermacher/Qwen3-VL-12B-Instruct-Brainstorm20x-GGUF](https://huggingface.co/mradermacher/Qwen3-VL-12B-Instruct-Brainstorm20x-GGUF)
+    **Model Type:** Multimodal Instruction-Tuned Vision-Language Model
+    **Architecture:** Qwen3-VL with Brainstorm 20x adapter (expanded to 12B parameters, 55 layers)
+    **Context Length:** Up to 256K tokens (extendable to 1M)
+    **Languages:** English (with strong multilingual OCR support)
+    **Key Features:**
+    - Advanced visual understanding: Recognizes objects, scenes, text, and spatial relationships with high precision.
+    - Long-form reasoning: Excels in complex, multi-step tasks including STEM, coding, and creative writing.
+    - Enhanced multimodal generation: Generates rich, detailed descriptions from images and videos; supports image-to-code (HTML/CSS/JS/Draw.io).
+    - Brainstorm 20x Enhancement: A proprietary adapter that significantly improves prose quality, coherence, emotional engagement, and creative depth—while maintaining strong instruction-following accuracy.
+    - Full support for long documents, books, and extended video analysis with second-level indexing.
+    - Integrates with GUI agents for PC/mobile task automation.
+
+    **Use Cases:**
+    - Creative writing, story generation, and narrative design
+    - Image and video analysis with rich contextual description
+    - Visual coding and interface design from screenshots
+    - Long-context reasoning and research
+    - AI-assisted brainstorming and ideation
+    - Advanced agent-based workflows
+
+    **Quantization Support:** GGUF (Q2_K to Q8_0), GPTQ, EXL2, AWQ, HQQ — with full support for **mmproj** quant (Q8, F16, BF16, F32) required for full functionality.
+
+    **Recommended Tools:** KoboldCpp, oobabooga/text-generation-webui, SillyTavern, or any GGUF-compatible inference engine.
+
+    **License:** Apache 2.0
+
+    > 📌 *Note: This is a fine-tuned, enhanced version of Qwen3-VL-8B-Instruct. The base model is officially released by Qwen. This variant includes the Brainstorm 20x expansion for deeper reasoning and richer output. Always use the full model with correct mmproj quantization for best results.*
+  overrides:
+    parameters:
+      model: Qwen3-VL-12B-Instruct-Brainstorm20x.Q4_K_S.gguf
+  files:
+    - filename: Qwen3-VL-12B-Instruct-Brainstorm20x.Q4_K_S.gguf
+      sha256: 42dbd1b00223c4d7030427ed4125c87329fc095ca9271826319148f8959dc875
+      uri: huggingface://mradermacher/Qwen3-VL-12B-Instruct-Brainstorm20x-GGUF/Qwen3-VL-12B-Instruct-Brainstorm20x.Q4_K_S.gguf