# Chinese Optimization

DeepSeek-R1-Distill-Qwen-14B is an optimized large language model with a parameter scale of 14B, released by DeepSeek AI. It is distilled from the Qwen architecture and offers multiple GGUF quantization versions to improve performance.

The AWQ quantized version of DeepSeek-R1-0528-Qwen3-8B, suitable for efficient inference in specific scenarios.

DeepSeek-R1T-Chimera is a text generation model developed based on tngtech's technology, focusing on efficient natural language processing tasks.

Qwen3-4B is a GGUF format model based on Qwen3-4B-Base, suitable for text generation tasks.

Qwen3-1.7B is a 1.7B-parameter open-source large language model released by Alibaba Cloud, based on the Transformer architecture, supporting various natural language processing tasks.

PEG is a model that achieves robust text retrieval through progressive learning, adjusting loss weights based on the difficulty levels of negative samples.

This model is converted from canopylabs/3b-zh-ft-research_release into GGUF format, suitable for Chinese text generation tasks.

GLM-4-32B-0414 is a large-scale language model developed by the THUDM team, based on the GLM architecture, suitable for various text generation tasks.

GLM-4-32B-0414 is a large-scale language model developed by THUDM, with 32 billion parameters, suitable for various text generation tasks.

A visual language model fine-tuned on the instancevid dataset based on Qwen2.5-VL-7B-Instruct, specializing in instance-level image description generation

A simplified 5-layer version of DeepSeek-V3 for lightweight tasks and rapid experimentation.

YOYO-O1-32B is a versatile large language model that integrates top-tier 32B reasoning and code models from the open-source community, built using SCE fusion technology.

This model is a GGUF format conversion of Google's gemma-3-27b-pt model via llama.cpp, suitable for local inference tasks.

This is a GGUF format model converted from the Qwen2-VL-7B-Captioner-Relaxed model, specifically designed for image-to-text tasks.

DeepSeek-R1-Distill-Llama-8B is an 8B-parameter large language model based on the Llama architecture, optimized through distillation training for text generation tasks.

This is the GPTQ 8-bit quantized version of the Llama-3.3-70B-Instruct model, fine-tuned and optimized for conversational reasoning tasks.

Qwen2-VL-7B is a vision-language model that provides GGUF format model files in multiple quantization versions, suitable for use with llama.cpp.

Llama-3.2-Taiwan-1B is a multilingual text generation model based on Meta's Llama-3.2-1B model, with special support for Chinese (Taiwan region) and multiple other languages.

GLM-Edge-1.5B-Chat is a 1.5 billion parameter-scale chat model based on the GLM architecture, suitable for Chinese dialogue scenarios.

SummLlama3.2-3B is a 3.2B-parameter summary generation model optimized based on the Llama3 architecture, offering multiple quantization versions to accommodate different hardware requirements.

MiniCPM-Embedding is an embedding model developed based on the MiniCPM-2B-sft-bf16 foundation model, specializing in retrieval tasks and supporting both Chinese and English.

LinFusion is a text-to-image generation model based on the Diffusers library, supporting the creation of high-quality images from text descriptions.

This is an open-source model based on the Apache-2.0 license. Specific details need to be added

This model is released under the Apache-2.0 license, with specific details currently unknown.

LLaVA model fine-tuned based on Meta-Llama-3-8B-Instruct and CLIP-ViT-Large-patch14-336, supporting image-to-text tasks

A multimodal model fine-tuned based on Meta-Llama-3-8B-Instruct and CLIP-ViT-Large-patch14-336, supporting image understanding and text generation

This model is a GGUF format version converted from xtuner/llava-llama-3-8b-v1_1, suitable for the llama.cpp framework, supporting image-text-to-text conversion tasks.

Open-source model based on Apache-2..0 license, specific functionalities depend on the actual model type

Chinese-LLaMA-2-7B is an extended Chinese version of Meta's Llama-2 model, optimized with an expanded Chinese vocabulary and incremental pre-training to enhance Chinese comprehension.

A Chinese T5-large model specialized in natural language transformation tasks, utilizing BertTokenizer and a Chinese character-level vocabulary.

This is a small BERT model based on the ConGen framework, designed to map sentences into a 512-dimensional vector space, suitable for tasks like semantic search.

A sentence embedding model based on ConGen, capable of mapping sentences to a 312-dimensional vector space, suitable for tasks like semantic search.

A versatile large language model capable of handling various natural language tasks

This model was pre-trained on a 250-million-word Chinese corpus using the MobileBERT architecture, with a training period of 15 days, completing 1 million iterations on a single A100 GPU.

DialoGPT is a dialogue generation model based on the GPT architecture, developed by Microsoft, specifically designed to generate natural and fluent conversational responses.

A sci-fi/fantasy-themed language model fine-tuned based on GPT-Neo 2.7B

DialoGPT is a dialogue generation pre-trained model released by Microsoft, based on the GPT architecture and specifically designed for multi-turn dialogue tasks.

DialoGPT is a dialogue generation model released by Microsoft, based on GPT-2, specifically designed for generating natural and fluent dialogue responses.