GLM-4.6-GPTQ-Int4-Int8Mix is a quantized version based on the zai-org/GLM-4.6 base model, which performs excellently in text generation tasks. This model achieves optimization of the model size through GPTQ quantization technology while maintaining good performance.

Natural Language ProcessingTransformers

This is the FP8 dynamic quantization version of the NVIDIA-Nemotron-Nano-9B-v2 model. Optimization is achieved by quantizing weights and activations to the FP8 data type, significantly reducing disk size and GPU memory requirements by approximately 50% while maintaining excellent text generation performance.

Natural Language ProcessingTransformersMultiple Languages

This is a 4.2B parameter Mixture of Experts model based on the GPT-OSS architecture. After pruning optimization, only 4 experts are retained, and it is converted to a quantized version in GGUF format. The model focuses on efficient inference and text generation tasks, using Q4_K_M quantization technology to reduce memory usage while maintaining performance.

Natural Language ProcessingGgufMultiple Languages

This is a specialized mixture of experts model with a scale of 6B parameters based on the GPT architecture. After pruning optimization, only 7 experts are retained, and it is converted into a quantized version in the GGUF format. The model uses Q8_0 quantization, significantly reducing memory usage and computational requirements while maintaining performance.

Natural Language ProcessingGgufMultiple Languages

This is the ik_llama.cpp imatrix quantized version of the zai-org/GLM-4.5 model. It uses advanced quantization techniques to reduce the model's storage space while maintaining high prediction accuracy. It is designed for performance and efficiency optimization in specific scenarios.

Natural Language ProcessingGguf

A quantized version of the Tencent Hunyuan 4B Instruction Model based on llama.cpp. Through quantization technology, the model can run more efficiently in resource-constrained environments while maintaining a certain level of performance and quality. It offers multiple quantization type options and supports online repackaging optimization.

Natural Language ProcessingGguf

rwkv7-7.2B-g0 is a multilingual text generation model with 7.2 billion parameters based on the RWKV7 architecture. It uses innovative quantization optimization techniques, supports multiple languages such as English, Chinese, Japanese, Korean, French, Arabic, Spanish, and Portuguese, and has efficient inference performance and a wide range of application scenarios.

Natural Language ProcessingGgufMultiple Languages

SmolLM3-3B-INT8-INT4 is a quantized version based on the HuggingFaceTB/SmolLM3-3B model. It uses torchao to implement 8-bit embedding, 8-bit dynamic activation, and 4-bit weight linear quantization. The model is converted to the ExecuTorch format and achieves high performance on the CPU backend through optimization, making it particularly suitable for mobile device deployment.

Natural Language ProcessingPytorch

A community version based on the Baidu ERNIE-4.5-21B-A3B model, implementing 8-bit quantization optimization using MLX, specifically optimized for Apple chip devices to improve performance and suitable for text generation tasks.

Natural Language ProcessingTransformersMultiple Languages

StableLM Zephyr 3B is a 3-billion-parameter instruction-tuned model trained on public datasets, synthetic datasets, and Direct Preference Optimization (DPO), delivering excellent performance.

Natural Language ProcessingGgufEnglish

A comprehensive re-fusion and remastered 32k context large language model based on the Dark Sapling V1 7B model, with significant improvements in quantization level optimization and performance

Natural Language ProcessingGgufEnglish

The Bigger Bodies Initiative is a character simulation optimization model developed by Playtime Co., built on Meta Llama-3.3-70B-Instruct. This model focuses on adaptive character simulation across different age groups and shows certain performance in multilingual processing. However, there are data penetration issues, and special attention should be paid to security protocols.

Natural Language ProcessingTransformersMultiple Languages

R1-AQA is an audio question answering model based on Qwen2-Audio-7B-Instruct, optimized through Group Relative Policy Optimization (GRPO) algorithm, achieving state-of-the-art performance in the MMAU benchmark.

MultimodalTransformers

Pensez is a bilingual French-English reasoning model that maximizes efficiency by significantly reducing training data. This model is built on Qwen 2.5 Instruct 7B and focuses on daily reasoning tasks and scientific problems. It uses optimization strategies such as concise reasoning and extended reasoning to improve performance.

Natural Language ProcessingTransformersMultiple Languages

LLaVA-Video-7B-Qwen2-TPO is a video understanding model based on LLaVA-Video-7B-Qwen2 with temporal preference optimization, demonstrating excellent performance across multiple benchmarks.

MultimodalTransformers

EAGLE is a new baseline method for fast decoding of large language models, which can significantly improve the generation efficiency while ensuring performance. The EAGLE series includes three versions: EAGLE-1, EAGLE-2, and EAGLE-3, achieving a 2 - 5.6 times increase in generation speed through continuous optimization.

Natural Language ProcessingTransformers

A fine-tuned model based on OpenAI CLIP ViT-L/14, achieving performance improvements through Geometric Parametrization (GmP), with special optimization for text encoding capabilities

MultimodalTransformers

HPC-Coder-v2-6.7b is a large language model fine-tuned on high-performance computing domain instruction datasets, focusing on typical HPC topics such as parallel computing, performance optimization, and accelerator porting.

Natural Language ProcessingTransformersMultiple Languages

SimPO is a simple preference optimization method that eliminates the need for reference rewards, aiming to enhance model performance by simplifying the preference optimization process.

Natural Language ProcessingTransformers

Excalibur-7b-DPO is a large language model based on the Excalibur-7b foundation model, fine-tuned with Direct Preference Optimization (DPO), focusing on improving dialogue quality and performance in visual application scenarios.

Natural Language ProcessingTransformers