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 the MXFP4_MOE quantized version of the MiniMax-M2-REAP-172B-A10B model, which is a memory-efficient compressed model. Through the REAP (Routing-Weighted Expert Activation Pruning) method, the model is compressed from 230B parameters to 172B parameters while maintaining performance, with a 25% reduction in size. It is suitable for resource-constrained environments, local deployment, and academic research.

Natural Language ProcessingGguf

VibeThinker-1.5B-GGUF is the GGUF quantized version of the WeiboAI/VibeThinker-1.5B model, quantized by MaziyarPanahi. This model is a text generation model with 1.5B parameters, supporting multiple quantization levels (from 2-bit to 8-bit), which facilitates efficient operation on various hardware.

Natural Language ProcessingGguf

This is a quantized version of the p-e-w/gpt-oss-20b-heretic model, quantized using the imatrix technology of llama.cpp. The model is a large language model with 20 billion parameters, offering a variety of quantization options ranging from high quality to low quality, with file sizes from 41.86GB to 11.52GB, suitable for different hardware conditions.

Natural Language ProcessingGguf

This is a large language model with 30 billion parameters trained on the Claude Sonnet 4.5 reasoning dataset. It is specifically optimized for high-difficulty reasoning tasks and performs excellently in the fields of coding and science.

Natural Language ProcessingTransformersEnglish

MiniMax-M2-REAP-162B-A10B is an efficiently compressed version of MiniMax-M2. It uses the REAP (Routing Weighted Expert Activation Pruning) method to reduce the model size by 30% while maintaining almost the same performance, compressing from 230B parameters to 162B parameters, significantly reducing the memory requirement.

Natural Language ProcessingTransformersEnglish

MiniMax-M2-REAP-172B-A10B is a memory-efficient compressed variant of MiniMax-M2. It uses the REAP expert pruning method, reducing the model size by 25% from 230B parameters to 172B parameters while maintaining almost the same performance.

Natural Language ProcessingTransformersEnglish

A large language model with 12B parameters fine-tuned based on Qwen 3, capable of handling 256k ultra-long contexts. It is specifically optimized for long text generation tasks. This model combines the foundation of the VLTO 8B model, Brainstorm 20x tuning, and NEO Imatrix dataset quantization technology, enabling it to generate high-quality long text content.

Natural Language ProcessingGgufMultiple Languages

Prima-24B is a large language model with 24 billion parameters, which is quantized and optimized through the GGUF format and specifically optimized for role-playing and creative writing tasks. This model is converted from the original Prima-24B model, using the Q4_K_M quantization level, which significantly reduces the model size and memory requirements while maintaining good performance.

Natural Language ProcessingGgufEnglish

Jan-v2-VL is a vision-language model with 8 billion parameters, designed specifically for long-term, multi-step tasks in real software environments (such as browsers and desktop applications). It combines language reasoning with visual perception, can follow complex instructions, maintain intermediate states, and recover from minor execution errors.

MultimodalGguf

Jan-v2-VL is a vision-language model with 8 billion parameters, designed to execute long-term, multi-step tasks in real software environments such as browsers and desktop applications. It tightly integrates language reasoning with visual perception, can follow complex instructions, maintain intermediate states, and recover from minor execution errors to achieve stable and continuous task completion.

MultimodalTransformersEnglish

Tarka-Embedding-350M-V1 is a text embedding model with 350 million parameters, capable of generating 1024-dimensional dense text representations. This model is optimized for downstream applications such as semantic similarity, search, and retrieval-augmented generation (RAG), supports multiple languages, and has the ability to handle long contexts.

Natural Language ProcessingSentence-transformers

Moondream 3 Preview HF is a re-implementation of the Moondream 3 (Preview) model based on the HuggingFace Transformers architecture specification, making it fully compatible with the Hugging Face ecosystem. It is a multimodal vision-language model that uses a Mixture of Experts (MoE) text backbone, with approximately 9 billion parameters and 2 billion active parameters.

MultimodalTransformers

This is an enhanced multimodal vision-language model based on the Qwen3-VL-8B-Thinking model. It is extended to 12B parameters through Brainstorm 20x technology and uses NEO Imatrix-enhanced GGUF quantization. The model has powerful image understanding, text generation, and multimodal reasoning capabilities, with significant improvements in visual perception, text quality, and creative scenarios.

MultimodalGgufMultiple Languages

NV-Reason-CXR-3B GGUF is a quantized version of the NVIDIA NV-Reason-CXR-3B vision-language model, optimized for edge device deployment. This is a model with 3 billion parameters, focusing on chest X-ray analysis. It has been converted to the GGUF format and quantized for efficient operation on mobile devices, desktops, and embedded systems.

MultimodalGgufEnglish

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 is a GGUF format version converted from the Qwen3-VL-30B-A3B-Instruct model, specifically optimized for llama.cpp. This model is a multimodal visual language model with 30 billion parameters, supporting image understanding and text generation tasks.

MultimodalGguf

Tarka-Embedding-150M-V1 is an embedding model with 150 million parameters, capable of generating 768-dimensional dense text representations. It is optimized for various downstream applications such as semantic similarity, search, and retrieval-augmented generation (RAG), focusing on capturing deep contextual semantics to support general text understanding across different domains.

Natural Language ProcessingSentence-transformers

Qwen3-Coder-480B-A35B-Instruct is a large code generation model with 480 billion parameters, supporting 8.5-bit quantization and optimized based on the MLX framework. This model is specifically designed for code generation tasks and can run efficiently on devices with sufficient memory.

Natural Language ProcessingMlxEnglish

Apollo-V0.1-4B-Thinking is a large language model based on 4B parameters, adopting a chain of thought reasoning architecture and specifically optimized for reasoning tasks. This model offers multiple quantization versions and can run efficiently in different hardware environments.

Natural Language ProcessingGguf