# Open-source Large Model

SWE-agent-LM-32B is an open-source software engineering language model, fine-tuned based on Qwen/Qwen2.5-Coder-32B-Instruct, and is specifically designed for software engineering tasks.

dots.llm1 is a large-scale MoE model that activates 14 billion parameters out of a total of 142 billion parameters, and its performance is comparable to that of the state-of-the-art models.

This is the GGUF quantized version of Google Gemma 3B model, suitable for local deployment and inference.

This is the GGUF quantized version of Google Gemma 3B model, suitable for local inference and supports text generation tasks.

Quantized version of DeepSeek-R1-Distill-Qwen-32B, reducing memory requirements and improving computational efficiency through INT8 weight quantization and activation quantization

Reflection Llama-3.1 70B is an open-source large language model trained with 'reflection tuning' technology, capable of autonomously detecting reasoning errors and correcting its approach.

Tarsier-7b is an open-source large-scale video-language model from the Tarsier series, specializing in generating high-quality video descriptions with excellent general video understanding capabilities.

Wizard Lake 7B is a fusion model combining the next-generation WizardLM 2 7B model with a customized DolphinLake model, delivering outstanding performance.

WizardLM-2 7B is a highly efficient large language model developed by Microsoft AI team, based on the Mistral-7B architecture, excelling in multilingual, reasoning, and agent tasks.

CogVLM is a powerful open-source vision-language model supporting tasks like object detection and visual question answering, featuring 4-bit precision quantization.

Aya 101 is a large-scale multilingual generative language model supporting instructions in 101 languages, outperforming similar models in various evaluations.

Emollama-chat-7b is part of the EmoLLMs project and is the first open-source large language model series with instruction-following capability, focusing on comprehensive sentiment analysis.

supermario-v2 is a merged model based on Mistral-7B-v0.1, combining three different models using the DARE_TIES method, with strong text generation capabilities.

Mixtral-8x7B is a pre-trained generative sparse mixture of experts model that outperforms Llama 2 70B on most benchmarks.

Yi-34B-Chat is a bilingual-optimized large language model developed by 01.AI, excelling in language understanding, commonsense reasoning, and reading comprehension, supporting both Chinese and English interactions.

koOpenChat-sft is a Korean dialogue model optimized based on OpenChat3.5, supporting ChatML and Alpaca format instruction interactions.

A 6.9 billion parameter dialogue-specific language model developed by Together in collaboration with multiple AI research institutions, trained on the RedPajama-Data-1T dataset and enhanced with dialogue capabilities through fine-tuning with OASST1 and Dolly2 data

FLAN-T5 is an instruction-fine-tuned language model based on T5, supporting 60+ languages, achieving stronger performance through fine-tuning on 1000+ tasks with the same parameter count