JanusCoder-8B is an open-source code intelligence foundation model built on Qwen3-8B, aiming to establish a unified visual programming interface. This model is trained on JANUSCODE-800K (the largest multimodal code corpus to date) and can handle various visual programming tasks, including data visualization, interactive web UI, and code-driven animation.

MultimodalTransformers

Lapa LLM v0.1.2 is an open-source large language model for Ukrainian language processing developed based on Gemma-3-12B. It focuses on natural language processing tasks in Ukrainian and shows excellent performance in Ukrainian language processing.

MultimodalTransformersMultiple Languages

Bee-8B is an advanced fully open-source multimodal large language model that focuses on data quality and aims to narrow the performance gap with proprietary models. By using the high-quality Honey-Data-15M corpus and the advanced data processing pipeline HoneyPipe, it demonstrates excellent performance in complex reasoning and other aspects.

MultimodalTransformersEnglish

KAT-Dev-72B-Exp is an open-source large language model with 72 billion parameters for software engineering tasks, achieving an accuracy of 74.6% in the SWE-Bench Verified benchmark test. It is an experimental reinforcement learning version of the KAT-Coder model.

Natural Language ProcessingTransformers

OpenLLaMA 3B v2 is an open-source large language model based on the Transformer architecture, with 3 billion parameters. This model uses the MIT license and is mainly used for English text generation tasks, supporting various application scenarios such as chatbots.

Natural Language ProcessingPeftEnglish

Lapa LLM 12B PT is an open-source large language model developed based on Google Gemma 3-12B, specifically optimized for Ukrainian language processing. It is developed by research teams from multiple Ukrainian universities and performs excellently in Ukrainian language processing, with efficient instruction fine-tuning and multimodal capabilities.

MultimodalTransformersOther

VLA-Adapter is a micro vision-language-action model trained on Libero-Spatial. It adopts the Prismatic-VLM architecture and only uses Qwen2.5-0.5B as the large language model backbone. This model outperforms open-source VLA models with larger parameter scales in robot benchmarks, achieving high-performance vision-language-action understanding and execution.

MultimodalSafetensorsEnglish

MiMo Audio is an audio language model developed by Xiaomi, which demonstrates strong few-shot learning ability through large-scale pre-training. This model breaks through the limitations of traditional models that rely on fine-tuning for specific tasks and performs excellently in tasks such as speech intelligence and audio understanding, reaching an advanced level among open-source models.

Audio ProcessingSafetensors

MiMo Audio is an audio language model based on large-scale pre-training, achieving SOTA performance among open-source models in speech intelligence and audio understanding benchmark tests. This model demonstrates strong few-shot learning ability and can generalize to tasks not included in the training data, supporting various audio tasks such as speech conversion, style transfer, and speech editing.

Audio ProcessingSafetensors

Isaac-0.1 is the first open-source vision-language model launched by PerceptronAI. It has 2 billion parameters and is specifically designed for real-world applications. This model performs excellently in multimodal understanding and spatial reasoning, achieving performance that meets or exceeds models more than 50 times larger, setting a new efficiency standard.

MultimodalTransformers

LLaVA-OneVision-1.5 is a series of fully open-source large multimodal models that achieve advanced performance at a lower cost by training on native resolution images. This model demonstrates excellent performance in multiple multimodal benchmark tests, surpassing competitors such as Qwen2.5-VL.

MultimodalTransformers

LLaDA-MoE is a new type of mixture-of-experts language model constructed based on the diffusion principle. It is the first open-source MoE diffusion large language model, pre-trained from scratch on approximately 20 trillion tokens. It has a total of 7 billion parameters, and only 1.4 billion parameters are activated during inference. It performs excellently in tasks such as code generation and mathematical reasoning.

Natural Language ProcessingTransformers

MiMo is a series of large language models trained from scratch by Xiaomi specifically for inference tasks. Through optimizing pre-training and post-training strategies, it demonstrates excellent performance in mathematical and code reasoning tasks. The project has open-sourced multiple versions with a 7B parameter scale, including the base model, SFT model, and RL model.

Natural Language ProcessingTransformers

InternVL3.5 is a new member of the open-source multimodal model family, significantly improving the versatility, reasoning ability, and reasoning efficiency of the InternVL series. It supports new functions such as GUI interaction and reaches the advanced level of open-source multimodal large language models.

MultimodalTransformersOther

Seed-OSS-36B-Instruct is a large language model developed by ByteDance-Seed, with 36 billion parameters and released under the Apache 2.0 open-source license. This model is built on the transformers library, supports vllm and mlx technology optimization, and has undergone 8-bit quantization specifically for Apple Silicon chips, providing efficient text generation capabilities.

Natural Language ProcessingTransformers

Seed-OSS is an open-source large language model series developed by the ByteDance Seed team. It has powerful long context processing, reasoning, and agent interaction capabilities. Trained with only 12T tokens, it performs excellently in multiple public benchmark tests and supports native long context processing of up to 512K.

Natural Language ProcessingTransformers

Seed-OSS-36B-Instruct is a large language model with 36 billion parameters developed by ByteDance and is open-sourced under the Apache-2.0 license. This model is specifically optimized for instruction following tasks, supporting text generation and dialogue functions, and has powerful understanding and generation capabilities.

Natural Language ProcessingGguf

Seed-OSS-36B-Instruct is a text generation model developed by ByteDance. It is based on a large language model architecture with a 36B parameter scale and is specifically optimized for instruction-following tasks. This model supports both English and Chinese and uses the Apache-2.0 open-source license. It can be efficiently deployed through the vllm and mlx inference frameworks.

Natural Language ProcessingMlxMultiple Languages

Seed-OSS-36B-Instruct is a large language model with 36 billion parameters developed by ByteDance. It is optimized based on the MLX framework and focuses on text generation tasks. This model supports both English and Chinese, and uses the Apache 2.0 open-source license. It has powerful instruction-following and content generation capabilities.

Natural Language ProcessingMlxMultiple Languages

Seed-OSS is an open-source large language model series developed by the ByteDance Seed team. It has powerful long context processing, reasoning, agent interaction capabilities, and general performance. This model was trained with only 12T tokens and performed excellently in multiple public benchmark tests.

Natural Language ProcessingTransformersMultiple Languages