Qwen3-VL is the latest generation of vision-language model launched by Alibaba, with significant improvements in text understanding, visual perception, spatial understanding, video analysis, and agent interaction. This model supports multimodal input and has powerful reasoning ability and long context processing ability.

MultimodalTransformers

Qwen3-VL-30B-A3B-Instruct-AWQ is a quantized version of the Qwen3-VL series, which has powerful visual language processing capabilities and supports tasks such as image understanding, video analysis, and multimodal reasoning. The model has significant improvements in text understanding, visual perception, spatial understanding, and long context processing.

MultimodalTransformers

The Ming-flash-omni preview version is a multimodal large model built on the Ling-Flash-2.0 sparse mixture of experts (MoE) architecture, with a total of 100B parameters and only 6B parameters activated per token. This model is a comprehensive upgrade based on Ming-Omni, showing significant improvements in multimodal understanding and generation, especially in speech recognition, image generation, and segmentation editing.

MultimodalDiffusersEnglish

DIMI Arabic OCR v2 is an optical character recognition model specifically optimized for Arabic text recognition. It is developed based on fine-tuning Qwen2.5-VL-7B-Instruct and has significant improvements in handling texts with dense diacritics compared to the v1 version.

MultimodalPeftArabic

Wan2.2 is a significant upgrade version of the basic video model. It introduces the Mixture of Experts (MoE) architecture, incorporates carefully curated aesthetic data, and is trained on larger datasets to enhance the ability to generate complex motions. This model supports generating 5-second videos with 480P and 720P resolutions, and has significant improvements in video generation quality and performance.

MultimodalDiffusers

Granite-4.0-H-Small is a long-context instruction model with 32 billion parameters developed by IBM, fine-tuned based on Granite-4.0-H-Small-Base. This model combines open-source instruction datasets and internal synthetic datasets and is developed using techniques such as supervised fine-tuning, reinforcement learning alignment, and model merging, with significant improvements in instruction following and tool invocation capabilities.

Natural Language ProcessingTransformers

GLM-4.6-AWQ is a quantized version based on the zai-org/GLM-4.6 base model, with efficient text generation capabilities. Compared with GLM-4.5, this model has significant improvements in aspects such as context window, encoding performance, reasoning ability, and agent functions.

Natural Language ProcessingTransformers

GLM-4.6 is a new-generation text generation model launched by Zhipu AI. Compared with GLM-4.5, it has significant improvements in multiple aspects, including a longer context window, better coding performance, and stronger reasoning ability. This model performs excellently in multiple public benchmark tests and has a competitive advantage compared with leading models at home and abroad.

Natural Language ProcessingTransformersMultiple Languages

GLM-4.6 is a new-generation large language model developed by Zhipu AI. Compared with GLM-4.5, it has significant improvements in context processing, coding ability, and reasoning performance. This model supports a context length of 200K and performs excellently in multiple public benchmark tests, especially having competitive advantages in code generation, reasoning, and agent tasks.

Natural Language ProcessingTransformersMultiple Languages

GLM-4.6 is a new-generation text generation model launched by Zhipu AI. Compared with GLM-4.5, it has achieved significant improvements in context processing, encoding performance, reasoning ability, etc. It supports a context length of 200K and has stronger agent capabilities and refined writing abilities.

Natural Language ProcessingTransformersMultiple Languages

GLM-4.6-FP8 is a new-generation text generation model launched by Zhipu AI. It has significant improvements in multiple key aspects compared to GLM-4.5. This model has a longer context window, excellent coding performance, advanced reasoning ability, and more powerful agent functions, enabling it to handle more complex tasks.

Natural Language ProcessingTransformersMultiple Languages

Hermes 4 14B is a cutting-edge hybrid mode reasoning model developed by Nous Research based on Qwen 3 14B. It has significant improvements in mathematics, code, STEM, logical reasoning, creativity, and format-fidelity output, while maintaining the quality of a general assistant and broad neutral alignment.

Natural Language ProcessingTransformersEnglish

DeepSeek-V3.1 is a large language model that is post-trained based on DeepSeek-V3.1-Base. It has a 128K context length and supports mixed thinking modes, intelligent tool invocation, and code proxy functions. This model performs excellently in multiple benchmark tests, especially showing significant improvements in mathematical reasoning, code generation, and search agent tasks.

Natural Language ProcessingTransformersMultiple Languages

Qwen3-42B-A3B-2507 is a large language model with 42B parameters based on the Qwen3 architecture. It performs excellently in code generation and text generation tasks, supports multilingual processing, and has enhanced reasoning ability. This version has been specially optimized and shows performance improvements compared to previous versions in multiple benchmark tests.

Natural Language ProcessingMlxMultiple Languages

Jet-Nemotron is a new family of hybrid architecture language models that surpasses state-of-the-art open-source full-attention language models such as Qwen3, Qwen2.5, Gemma3, and Llama3.2, while achieving significant efficiency improvements - with up to 53.6x acceleration in generation throughput on H100 GPUs.

Natural Language ProcessingTransformersEnglish

A text-to-image generation model based on Stable Diffusion XL, focusing on generating high-quality and detailed images. It is particularly good at anime, semi-realistic, and realistic styles, with significant improvements in hand details and picture brightness.

Computer VisionDiffusersEnglish

Qwen3-4B-Instruct-2507 is a large language model with 4 billion parameters launched by Alibaba Cloud. Based on the Unsloth Dynamic 2.0 quantization technology, it has significant improvements in general capabilities, long-tail knowledge coverage, user preference alignment, and long context understanding.

Natural Language ProcessingTransformers

Qwen3-4B-Instruct-2507 is an updated version of Qwen3-4B in non-thinking mode, with significant improvements in general capabilities, long-tail knowledge coverage, user preference alignment, and long context understanding. It uses the Unsloth Dynamic 2.0 quantization technology and performs excellently in accuracy.

Natural Language ProcessingTransformers

Qwen3-4B-Instruct-2507-AWQ is an updated version of Qwen3-4B in non-thinking mode, with significant improvements in general capabilities, long-tail knowledge coverage, user preference alignment, and long context understanding. It can be used in various scenarios such as text generation, providing users with high-quality language interaction experiences.

Natural Language ProcessingTransformers

Moirai 2.0 is a decoder-only general time series prediction Transformer model that has been pre-trained on multiple high-quality datasets, including GIFT-Eval, a subset of the Chronos dataset, synthetic time series, and Salesforce's internal operational data. Compared with the first version, significant improvements have been made in aspects such as the loss function, prediction method, and data processing.

Scientific ComputingSafetensors