Gemma-UA-Cardio consists of two professional large language models specifically designed for the Ukrainian cardiology field. They are based on Google's Gemma and MedGemma models and have undergone two-stage fine-tuning (language adaptation and domain specialization). They can accurately provide cardiovascular medical information and answer clinical questions in Ukrainian, serving as auxiliary tools for medical professionals.

Natural Language ProcessingTransformersOther

Gemma 3n E4B IT is a lightweight multimodal open model launched by Google, built based on the same research as the Gemini model. This model supports text, audio, and visual inputs and is suitable for various tasks. It uses the MatFormer architecture to achieve efficient parameter utilization.

MultimodalTransformers

Gemma 3 is a lightweight open-source multimodal model launched by Google. This version is an instruction-tuned quantization-aware training model with 12B parameters, which has been converted to the MXFP4 format of the MLX framework. It supports text and image input and generates text output, with a 128K context window and support for over 140 languages.

MultimodalMlx

Qsana-coder-base is a small language model (SLM) specifically designed for conceptual creativity in coding basics (Python, pseudo-code). It does not generate executable production code but rather generates coding logic snippets for educational and rapid prototyping scenarios. Its main goal is to stimulate the logical thinking and conceptual understanding of beginners.

Natural Language ProcessingTransformersFrench

This is a housing condition classifier fine-tuned based on the Google ViT-base model, which can classify housing images into four categories: good, unknown, dilapidated, and medium. The model was trained on a dataset of 935 housing images, and the accuracy on the validation set reached 81.2%.

Computer VisionTransformersEnglish

This is the quantized version of kldzj's GPT-OSS-120B-Heretic-v2 model, which is quantized using the imatrix technology of llama.cpp. The model significantly reduces the storage and computing resource requirements through quantization technology while maintaining good model performance, facilitating deployment and operation on various devices.

Natural Language ProcessingGguf

This is a Dutch patient simulation model fine-tuned based on Google's Gemma 2 9B model, specifically designed for medical education scenarios. The model always responds in a fixed JSON structure, simulating a patient 'Maria' who has just completed surgery and is in the recovery process, supporting nursing students to exercise communication and clinical reasoning skills in a safe environment.

Natural Language ProcessingGgufOther

This is an experimental mixed quantization version based on the Qwen3 VL 8B model. It uses the MXFP4_MOE technology combined with high-precision weights to achieve a smaller file size and higher inference speed while maintaining accuracy close to Q8. This model explores mixed quantization methods and achieves a good balance between accuracy loss and performance.

MultimodalSafetensorsOther

This is an English-Spanish translation model fine-tuned on the Amazon dataset based on the Google mT5-small model, specifically designed for text abstract generation tasks. The model achieved scores of Rouge1: 16.44 and Rouge2: 8.04 on the evaluation set.

Natural Language ProcessingTransformers

Prima-24B is a large language model with 24 billion parameters, which is quantized and optimized through the GGUF format and specifically optimized for role-playing and creative writing tasks. This model is converted from the original Prima-24B model, using the Q4_K_M quantization level, which significantly reduces the model size and memory requirements while maintaining good performance.

Natural Language ProcessingGgufEnglish

Olmo 3 32B Think SFT is an autoregressive language model based on the Transformer architecture, which performs excellently in long-chain thought reasoning and is particularly good at handling complex reasoning tasks such as mathematics and coding. This model was pre-trained on the Dolma 3 dataset and supervised fine-tuned on the Dolci dataset.

Natural Language ProcessingTransformersEnglish

FLAN-T5 Small is an instruction fine-tuned text generation model developed by Google, based on the small version of the T5 architecture. This model has undergone instruction fine-tuning on various NLP tasks and can understand and execute various text generation instructions.

Natural Language ProcessingSafetensorsEnglish

This is a text generation model fine-tuned based on the Google Gemma-3-4B-IT model, specifically designed to generate diverse and relevant search queries according to the metadata of demo templates. This model uses the LoRA adapter for efficient fine-tuning and is a key component of the structured document field adaptive dense retrieval framework.

Natural Language ProcessingSafetensorsEnglish

Chandra is a high-precision OCR model that can convert images and PDFs into structured outputs, such as Markdown, HTML, and JSON, while retaining detailed layout information. It supports more than 40 languages and is good at handling complex document elements.

MultimodalTransformersEnglish

This is a text generation model optimized using PEFT (Parameter-Efficient Fine-Tuning) and LoRA techniques based on the Google Gemma-2B-IT base model. The model is optimized for specific tasks while maintaining the capabilities of the base model through parameter-efficient methods.

Natural Language ProcessingPeft

These are various GGUF format quantized versions of the Google embeddinggemma-300m base model, optimized for efficient deployment and use in different scenarios. This model focuses on sentence embedding and similarity calculation tasks.

Natural Language ProcessingGgufEnglish

This project performs 4-bit quantization on the MiniMax-M2 model using the DWQ (Dynamic Weight Quantization) method with the help of the mlx-lm library. This model is a lightweight version of MiniMax-M2, significantly reducing the model size while maintaining good performance.

Natural Language ProcessingMlx

Kimi Linear is an efficient hybrid linear attention architecture that outperforms traditional full attention methods in short context, long context, and reinforcement learning scenarios. It optimizes attention computation through the Kimi Delta Attention (KDA) mechanism, significantly improving performance and hardware efficiency, and is particularly good at handling long context tasks up to 1 million tokens.

Natural Language ProcessingTransformers

This model is an instance segmentation model fine-tuned on the qubvel-hf/ade20k-mini dataset based on facebook/mask2former-swin-tiny-coco-instance. This model is specifically optimized for the scene understanding task in the ADE20K-mini dataset and shows good performance in the instance segmentation task.

Computer VisionTransformers

This is the MXFP4_MOE quantized version of the Qwen3-Yoyo-V4-42B-A3B-Thinking-TOTAL-RECALL model, specifically optimized for text generation tasks. The model is based on a large language model architecture with 42B parameters. Through quantization technology, it reduces the computational and storage requirements while maintaining good performance.

Natural Language ProcessingGguf