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
@@ -23023,3 +23023,42 @@
     - filename: Evilmind-24B-v1.i1-Q4_K_M.gguf
       sha256: 22e56c86b4f4a8f7eb3269f72a6bb0f06a7257ff733e21063fdec6691a52177d
       uri: huggingface://mradermacher/Evilmind-24B-v1-i1-GGUF/Evilmind-24B-v1.i1-Q4_K_M.gguf
+- !!merge <<: *llama3
+  name: "lmunit-llama3.1-70b-i1"
+  urls:
+    - https://huggingface.co/mradermacher/LMUnit-llama3.1-70b-i1-GGUF
+  description: |
+    **Model Name:** LMUnit-llama3.1-70b
+    **Base Model:** Llama-3.1-70B-Instruct
+    **Developer:** Contextual AI
+    **Task:** Fine-grained natural language evaluation using unit tests
+    **Language:** English
+
+    **Description:**
+    LMUnit is a high-performance language model fine-tuned for precise, criterion-based evaluation of LLM responses. It takes a prompt, response, and a natural language unit test as input, then returns a continuous score (1–5) indicating how well the response satisfies the test criteria. Trained with multi-objective learning, synthetic data, and importance weighting, it achieves state-of-the-art performance across evaluation benchmarks like FLASK, BiGGen Bench, and RewardBench, with near-human alignment (93.5% accuracy on RewardBench). Ideal for detailed, reliable assessment of response quality in research, benchmarking, and model development.
+
+    **Key Features:**
+    - Optimized for fine-grained, criteria-driven evaluation
+    - High alignment with human preferences
+    - Superior performance on FLASK, BiGGen Bench, and LFQA
+    - Based on Llama-3.1-70B-Instruct (original, non-quantized version)
+
+    **Use Case:** Evaluating response accuracy, coherence, and adherence to specific criteria in complex or nuanced tasks.
+
+    **Citation:**
+    ```bibtex
+    @inproceedings{saadfalcon2025lmunit,
+      title={{LMUnit}: Fine-grained Evaluation with Natural Language Unit Tests},
+      author={Jon Saad-Falcon and Rajan Vivek and William Berrios and Nandita Shankar Naik and Matija Franklin and Bertie Vidgen and Amanpreet Singh and Douwe Kiela and Shikib Mehri},
+      booktitle={Findings of the Association for Computational Linguistics: EMNLP 2025},
+      year={2025},
+      url={https://arxiv.org/abs/2412.13091}
+    }
+    ```
+  overrides:
+    parameters:
+      model: LMUnit-llama3.1-70b.i1-Q4_K_M.gguf
+  files:
+    - filename: LMUnit-llama3.1-70b.i1-Q4_K_M.gguf
+      sha256: 4f2cff716b66a5234a1b9468b34ac752f0ca013fa31a023f64e838933905af57
+      uri: huggingface://mradermacher/LMUnit-llama3.1-70b-i1-GGUF/LMUnit-llama3.1-70b.i1-Q4_K_M.gguf