Qwen3-32B-AWQ

Run locally Apple devices with Mirai

Type

Type

Local

From

From

Alibaba

Quantisation

Quantisation

uint4

Precision

Precision

No

Size

Size

32B

Source

Source

Hugging Face Logo

Qwen3-32B-AWQ is a 32.8 billion parameter language model and the latest generation in the Qwen series, featuring a 4-bit AWQ quantization. It uniquely supports seamless switching between thinking mode for complex logical reasoning, mathematics, and coding, and non-thinking mode for efficient general-purpose dialogue, all within a single model. The model demonstrates significant improvements in reasoning capabilities, human preference alignment for creative writing and role-playing, agent capabilities for tool integration, and multilingual support across 100+ languages and dialects. Built on extensive pretraining and post-training, Qwen3-32B natively supports context lengths of 32,768 tokens and can extend to 131,072 tokens using YaRN rope scaling. The quantized version maintains strong performance relative to its full-precision counterpart while offering improved efficiency for deployment.

1
Choose framework
2
Run the following command to install Mirai SDK
SPMhttps://github.com/trymirai/uzu-swift
3
Set Mirai API keyGet API Key
4
Apply code
Loading...

Qwen3-32B-AWQ

Run locally Apple devices with Mirai

Type

Local

From

Alibaba

Quantisation

uint4

Precision

float16

Size

32B

Source

Hugging Face Logo

Qwen3-32B-AWQ is a 32.8 billion parameter language model and the latest generation in the Qwen series, featuring a 4-bit AWQ quantization. It uniquely supports seamless switching between thinking mode for complex logical reasoning, mathematics, and coding, and non-thinking mode for efficient general-purpose dialogue, all within a single model. The model demonstrates significant improvements in reasoning capabilities, human preference alignment for creative writing and role-playing, agent capabilities for tool integration, and multilingual support across 100+ languages and dialects. Built on extensive pretraining and post-training, Qwen3-32B natively supports context lengths of 32,768 tokens and can extend to 131,072 tokens using YaRN rope scaling. The quantized version maintains strong performance relative to its full-precision counterpart while offering improved efficiency for deployment.

1
Choose framework
2
Run the following command to install Mirai SDK
SPMhttps://github.com/trymirai/uzu-swift
3
Set Mirai API keyGet API Key
4
Apply code
Loading...