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
@@ -23203,3 +23203,45 @@
     - filename: Spiral-Qwen3-4B-Multi-Env.Q4_K_M.gguf
       sha256: e91914c18cb91f2a3ef96d8e62a18b595dd6c24fad901dea639e714bc7443b09
       uri: huggingface://mradermacher/Spiral-Qwen3-4B-Multi-Env-GGUF/Spiral-Qwen3-4B-Multi-Env.Q4_K_M.gguf
+- !!merge <<: *qwen3
+  name: "moonshotai.kimi-k2-thinking"
+  urls:
+    - https://huggingface.co/DevQuasar/moonshotai.Kimi-K2-Thinking-GGUF
+  description: |
+    **Model Name:** Kimi-K2-Thinking
+    **Repository:** [moonshotai/Kimi-K2-Thinking](https://huggingface.co/moonshotai/Kimi-K2-Thinking)
+    **License:** Modified MIT
+    **Architecture:** Mixture-of-Experts (MoE)
+    **Context Length:** 256K tokens
+    **Total Parameters:** 1T (activated: 32B)
+    **Quantization:** Native INT4 (quantization-aware trained for lossless performance)
+    **Pipeline Tag:** `text-generation`
+
+    ### 🌟 Description:
+    Kimi-K2-Thinking is a state-of-the-art open-source reasoning and agentic AI model developed by Moonshot AI. Designed for deep, multi-step reasoning and autonomous tool use, it excels in complex tasks requiring extended thought chains, such as research, coding, and long-horizon problem solving.
+
+    With a massive 256K context window and an MoE architecture (384 experts, 8 selected per token), it maintains coherent, goal-driven behavior across 200–300 sequential tool calls—surpassing previous models that degrade after 30–50 steps. It achieves **lossless INT4 quantization** via Quantization-Aware Training (QAT), enabling up to **2x faster inference** with minimal performance drop.
+
+    ### 🔍 Key Strengths:
+    - **Superior Reasoning & Tool Orchestration:** Leads benchmarks like *Humanity’s Last Exam (HLE)*, *SWE-bench*, *BrowseComp*, and *AIME25*.
+    - **Long-Horizon Agency:** Sustains stable, high-quality performance in multi-step workflows.
+    - **Efficient Deployment:** Optimized for vLLM, SGLang, and KTransformers with support for OpenAI-compatible APIs.
+    - **Open Access:** Full model weights available under a permissive license.
+
+    ### 🛠 Use Cases:
+    - Long-form reasoning & research agents
+    - Autonomous coding & debugging
+    - Multi-step web research & data gathering
+    - Tool-augmented chatbots and personal assistants
+
+    > 🔗 **Try it**: Access via [Kimi Platform](https://www.kimi.com) or deploy locally using vLLM.
+    > 📚 Learn more: [Tech Blog](https://moonshotai.github.io/Kimi-K2/thinking.html)
+
+    *Note: The GGUF version (e.g., `DevQuasar/moonshotai.Kimi-K2-Thinking-GGUF`) is a quantized derivative. Use the original `moonshotai/Kimi-K2-Thinking` for full capabilities and documentation.*
+  overrides:
+    parameters:
+      model: moonshotai.Kimi-K2-Thinking.Q4_K_M-00001-of-00053.gguf
+  files:
+    - filename: moonshotai.Kimi-K2-Thinking.Q4_K_M-00001-of-00053.gguf
+      sha256: ddfd84f484a1f548121374a6d437299fe4c3355118c98003c6efd0ad17cbcbd6
+      uri: huggingface://DevQuasar/moonshotai.Kimi-K2-Thinking-GGUF/moonshotai.Kimi-K2-Thinking.Q4_K_M-00001-of-00053.gguf