MERaLiON-SER-v1 is a parameter-efficient multilingual speech emotion recognition model that can simultaneously predict 7 discrete emotion categories and continuous emotion dimension values. This model is built on the Whisper-Medium encoder and adopts the LoRA adaptation technology, showing excellent performance in the multilingual speech emotion recognition task.

Audio ProcessingTransformersMultiple Languages

This project provides creative writing control vectors for the GLM-4.6 model. Through the control vector technology, the model's performance in specific dimensions can be adjusted, such as honesty and Machiavellianism tendencies, communication style, etc.

Natural Language ProcessingTransformersEnglish

KaLM-Embedding-V2.5 is a general-purpose and lightweight embedding model. Through excellent training techniques and data, it achieves the best performance among models of the same scale and can even compete with models 3 - 26 times larger in scale. It supports multiple languages and various embedding dimension options.

Natural Language ProcessingSentence-transformersMultiple Languages

MiniCPM4.1-8B GGUF is an efficient large language model designed specifically for edge devices. It has carried out systematic innovations in four key dimensions: model architecture, training data, training algorithm, and inference system, achieving extreme efficiency improvement. It supports a context length of 65,536 tokens and a hybrid inference mode.

Natural Language ProcessingTransformersMultiple Languages

This is a curated collection of large language models in GGUF format for various minority languages. It supports multiple languages such as Russian, English, and German, covering multiple test dimensions including fact - finding, mathematical calculation, cultural humor, and moral filtering.

Natural Language ProcessingGgufMultiple Languages

Granite Guardian 3.2 3B - A800M is a fine-tuned model specifically designed to detect risks in prompts and responses. It can conduct risk detection based on multiple key dimensions and outperforms other open-source models of the same type in standard benchmark tests.

Natural Language ProcessingTransformers

MiniCPM4-8B is an efficient large language model designed specifically for edge devices. Through innovations in four dimensions: model architecture, training data, training algorithms, and inference systems, it achieves extreme efficiency improvements.

Natural Language ProcessingTransformersMultiple Languages

MiniCPM4 is an efficient large - language model designed specifically for edge devices. Through systematic innovation, it achieves extreme efficiency improvements in four key dimensions: model architecture, training data, training algorithm, and inference system.

Natural Language ProcessingTransformersMultiple Languages

MiniCPM4 is an efficient large language model designed specifically for edge devices. Through systematic innovation, it achieves extreme efficiency improvements in four dimensions: model architecture, training data, training algorithm, and inference system. It can achieve over 5 times faster generation speed on edge chips.

Natural Language ProcessingTransformersMultiple Languages

This is an improved version based on the merging of multiple 70B parameter models, primarily adjusting the embedding layer dimensions to match standard L3 70B model specifications.

Natural Language ProcessingTransformersEnglish

A sentence transformer model trained on the PubMed dataset, supporting multiple embedding dimensions, suitable for biomedical text processing.

Natural Language ProcessingSentence-transformers

A legal domain sentence embedding model fine-tuned based on ModernBERT, supporting multi-level dimensional output, suitable for legal text similarity calculation and information retrieval tasks.

Natural Language ProcessingSafetensorsEnglish

Granite Guardian 3.1 2B is a fine-tuned Granite 3.1 2B instruction model designed to detect risks in prompts and responses. It can perform risk detection based on multiple key dimensions listed in the IBM AI Risk Atlas.

Natural Language ProcessingTransformersEnglish

This is a sentence transformer model fine - tuned from sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2, which can map text to a 384 - dimensional vector space for tasks such as semantic similarity calculation.

Natural Language ProcessingSafetensors

Oryx-1.5-7B is a 7B-parameter model developed based on the Qwen2.5 language model, supporting a 32K token context window and specializing in efficiently processing visual inputs of arbitrary spatial dimensions and durations.

MultimodalSafetensorsMultiple Languages

SILMA Arabic Matryoshka Embedding Model 0.1 is an advanced Arabic text embedding model that uses innovative matryoshka embedding technology to optimize text representation in different dimensions, balancing speed, storage, and accuracy.

Natural Language ProcessingSentence-transformersMultiple Languages

An image inpainting model based on the Kolors foundation model, supporting Chinese prompts, excelling in comprehensive satisfaction and restoration trace dimensions

Computer VisionDiffusersMultiple Languages

The OpenSearch Neural Sparse Encoding Model v2 Distilled is an efficient learned sparse retrieval model designed for OpenSearch, capable of encoding queries and documents into 30,522-dimensional sparse vectors.

Natural Language ProcessingTransformersEnglish

This is an emotion prediction model fine-tuned based on DeBERTaV3-large, specifically designed to predict the emotion trajectory in written stories, including two dimensions: valence (emotional positivity or negativity) and arousal (emotional intensity). The output range is a continuous value between 0 and 1.

Natural Language ProcessingTransformersEnglish

This model is trained on a large dataset of manually scored IELTS essays and can automatically score IELTS essays across multiple dimensions with an accuracy of 0.82.

Natural Language ProcessingTransformersEnglish