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

This is a distilled model based on the Qwen3 30B A3B model, trained on the Claude Sonnet 4.5 high-reasoning-difficulty dataset. This model is specifically optimized for complex reasoning tasks and performs excellently in the fields of coding and science, with strong logical reasoning ability.

Natural Language ProcessingTransformersEnglish

This project is a quantized version based on the Qwen3-30B-A3B-Thinking-2507-Deepseek-v3.1-Distill-V2-FP32 model, providing various GGUF format files, including F16 full precision and multiple quantized versions (Q5_K_XL, Q4_K_XL, Q3_K_XL), to meet different hardware environments and usage requirements.

Natural Language ProcessingGguf

The W4A16 quantized version of Qwen3-30B-A3B-Thinking-2507, using LLM-Compressor for 4-bit weight and 16-bit activation quantization. Compared with the FP16 version, the memory usage is reduced by about 75%, and faster inference speed is achieved on supported hardware. This model has excellent reasoning ability, tool invocation ability, and long context support.

Natural Language ProcessingTransformers

This is a quantized version in GGUF format based on the Qwen3-30B-A3B-Thinking-2507 model, processed by abliterated and using the Q4_K_M quantization level. This model focuses on text generation tasks, supports Chinese and English, and has a parameter scale of 30B.

Natural Language ProcessingTransformersEnglish

Qwen3-30B-A3B-Thinking-2507 is a large language model with significant improvements in reasoning ability and general ability. It enhances the ability to understand long contexts and is suitable for highly complex reasoning tasks. This model has 30.5 billion parameters, of which 3.3 billion are activated, and supports long context processing of 262,144 tokens.

Natural Language ProcessingSafetensors

Qwen3-30B-A3B-Thinking-2507-AWQ is an efficient inference model quantized based on the Qwen/Qwen3-30B-A3B-Thinking-2507 base model. This model has significant improvements in inference tasks, general capabilities, and long context understanding, and is specifically optimized for highly complex reasoning tasks, supporting long context processing of 262,144 tokens.

Natural Language ProcessingTransformers

Qwen3-30B-A3B-Thinking-2507-GPTQ-Int8 is a quantized version based on Qwen/Qwen3-30B-A3B-Thinking-2507, with significant improvements in reasoning tasks, general capabilities, and long context understanding capabilities. It is suitable for highly complex reasoning tasks.

Natural Language ProcessingTransformers

This model is an MLX format conversion version of Qwen/Qwen3-30B-A3B-Thinking-2507, providing an efficient solution for text generation tasks. It uses 4-bit quantization technology to optimize performance and supports efficient operation on Apple Silicon devices.

Natural Language ProcessingMlx

Qwen3-30B-A3B-Thinking-2507-AWQ is a version after AWQ quantization processing based on the Qwen3-30B-A3B-Thinking-2507 model. This model has significantly improved in aspects such as reasoning ability, general ability, and long context understanding ability. It supports a context length of 256K and can be applied to various natural language processing tasks.

Natural Language ProcessingTransformers

Qwen3-30B-A3B-Thinking-2507 is a quantized large language model with enhanced reasoning ability, general ability, and long context understanding ability. This model adopts a mixture of experts architecture and performs excellently in complex tasks such as logical reasoning, mathematics, science, and coding. It supports long context processing of 262,144 tokens.

Natural Language ProcessingTransformers