Z-Image is an efficient basic image generation model with 6 billion parameters, specifically designed to address the efficiency and quality issues in the field of image generation. Its distilled version, Z-Image-Turbo, can reach or exceed the leading competitors with only 8 function evaluations. It can achieve sub-second inference latency on enterprise-level H800 GPUs and run on consumer-grade devices with 16G VRAM.

Computer VisionDiffusersMultiple Languages

Z-Image is a powerful and efficient image generation model with 6 billion parameters. It can effectively solve the problems in the fields of image generation in terms of efficiency, quality, and functional diversity, providing users with high-quality image generation and editing services.

Computer VisionDiffusersEnglish

The quantized GGUF version of Tencent's Hunyuan Video 1.5 model, specifically designed for text-to-video generation tasks, supports 720P HD video generation, and offers multiple quantization precision versions to optimize usage efficiency.

MultimodalGguf

This is the GGUF quantized version of allenai's Olmo-3-32B-Think model. It undergoes various quantization processes using the llama.cpp tool, aiming to enhance the model's performance and efficiency in specific environments. It offers multiple quantization options from Q2 to Q8 to meet different hardware configurations and performance requirements.

Natural Language ProcessingGgufEnglish

GLM-4.5-Iceblink-v2-106B-A12B-FP8 is a version based on the GLM-4.5-Iceblink-v2-106B-A12B model, using the most advanced mixture of experts quantization method for FP8 quantization. This model is specifically optimized for Ada, Hopper, or Blackwell series GPUs that support hardware FP8, significantly improving inference efficiency while maintaining high-quality output.

Natural Language ProcessingSafetensors

This is a quantized version based on the Huihui-MiroThinker-v1.0-8B model, specifically optimized for text generation tasks. It improves inference efficiency through quantization technology while maintaining the model's performance.

Natural Language ProcessingGguf

This is a version of the MiroThinker-v1.0-30B model quantized using MXFP4_MOE imatrix, implemented based on mradermacher's imatrix. This quantized model maintains the text generation ability of the original model while improving inference efficiency through quantization technology. It is suitable for various application scenarios that require efficient text generation.

Natural Language ProcessingGguf

This project provides a quantized version of the cerebras/MiniMax-M2-REAP-172B-A10B model, aiming to make knowledge accessible to the public. This is a large language model with 172 billion parameters, which has been optimized and quantized to reduce deployment costs and improve inference efficiency.

Natural Language ProcessingGguf

This is a quantized version of the kldzj/gpt-oss-120b-heretic model, quantized using llamacpp. It offers a variety of quantization types, including special formats such as BF16, Q8_0, and MXFP4_MOE, significantly improving the model's operating efficiency.

Natural Language ProcessingGguf

This is a mixed quantization version based on the Qwen3 VL 8B model, which uses MXFP4 quantization technology and mixed weights on the dense model. While maintaining accuracy close to Q8, this model achieves a smaller file size and higher inference speed, making it an excellent solution for balancing efficiency and accuracy.

MultimodalSafetensorsOther

This project is the result of merging pretrained language models using mergekit. It combines multiple pretrained models in the medical field, providing stronger support for medical Q&A tasks and effectively improving the accuracy and efficiency of answering medical questions.

Natural Language ProcessingSafetensorsEnglish

MiniMax-M2 AWQ - INT4 is a quantized version based on the MiniMax-M2 model. It uses INT4 quantization technology to significantly reduce memory usage and improve inference efficiency while ensuring performance. This model performs excellently in coding and agent tasks and has outstanding comprehensive performance.

Natural Language ProcessingTransformers

Aquif-3.5-Max-42B-A3B is a large language model with 42 billion parameters, which has undergone MXFP4_MOE quantization processing. It optimizes inference efficiency while maintaining high-quality text generation capabilities. This model is based on an advanced mixture-of-experts architecture and is suitable for various natural language processing tasks.

Natural Language ProcessingGguf

This project is the MXFP4_MOE quantized version of the Kanana 1.5-15.7B-A3B instruction model, which can improve the model's operating efficiency in specific scenarios. This model is optimized based on kakaocorp/kanana-1.5-15.7b-a3b-instruct, using a mixture-of-experts architecture and MXFP4 quantization technology.

Natural Language ProcessingGguf

Gelato-30B-A3B is a state-of-the-art (SOTA) model fine-tuned for GUI computer usage tasks, offering a quantized version to optimize deployment efficiency. This model is specifically designed to understand and process tasks related to graphical user interfaces.

MultimodalGguf

This is a fine-tuned Whisper speech recognition model based on the unsloth/whisper-large-v3-turbo architecture. It uses the Unsloth and Huggingface TRL libraries to achieve 2x training acceleration, significantly improving training efficiency.

Audio ProcessingTransformersEnglish

This is the GGUF quantized version of the NVIDIA ChronoEdit-14B-Diffusers model, specifically designed for image-to-video tasks. This model retains all the functions of the original model while optimizing deployment and runtime efficiency through the GGUF format.

Computer VisionGguf

Chronos-2 is a time series foundation model with 120 million parameters, supporting zero-shot prediction. It supports univariate, multivariate, and covariate-aware tasks within a single architecture, achieving state-of-the-art accuracy in zero-shot prediction across multiple benchmarks, and has extremely high inference efficiency.

Scientific ComputingSafetensors

Kimi Linear is a hybrid linear attention architecture that outperforms traditional full attention methods in various scenarios, including short and long contexts, as well as reinforcement learning extension mechanisms. It can effectively solve the problem of low efficiency of traditional attention mechanisms in long context tasks and bring more efficient solutions to fields such as natural language processing.

Natural Language ProcessingTransformers

Kimi Linear is an efficient hybrid linear attention architecture that outperforms traditional full attention methods in short context, long context, and reinforcement learning scenarios. It optimizes attention computation through the Kimi Delta Attention (KDA) mechanism, significantly improving performance and hardware efficiency, and is particularly good at handling long context tasks up to 1 million tokens.

Natural Language ProcessingTransformers