Qwen3-30B-A3B

Qwen

Qwen/Qwen3-30B-A3B

GenerationOpen weightsApache 2.0Verified metrics

View on Hugging Face ↗

About

Mixture-of-experts model with roughly 30B total parameters and about 3B active, trading memory for inference speed.

Metrics

Downloads

2,375,419

last 30 days

Hub likes

930

Parameters

30B

published total

How it compares in Generation

SizeQwen3-30B-A3B sits in 10-100B — larger than 73% of the 899 models in this category with a published size.

<1B53

1-10B444

10-100B334

>100B68

Details

Model ID: Qwen/Qwen3-30B-A3B
Developer: Qwen
Category: Generation
Access: Open weights
License: Apache 2.0
Source: Hugging Face
Data provenance: Metrics fetched per-model from the Hugging Face API.
Link: https://huggingface.co/Qwen/Qwen3-30B-A3B

More from Qwen

Qwen3-0.6BQwen/Qwen3-0.6BGeneration Qwen3-8BQwen/Qwen3-8BGeneration Qwen2.5-7B-InstructQwen/Qwen2.5-7B-InstructConversational Qwen2.5-0.5B-InstructQwen/Qwen2.5-0.5B-InstructConversational Qwen2.5-1.5B-InstructQwen/Qwen2.5-1.5B-InstructConversational Qwen3-4BQwen/Qwen3-4BGeneration

Other Generation models

gpt2openai-community/gpt2Generation Qwen3.6-35B-A3B-NVFP4nvidia/Qwen3.6-35B-A3B-NVFP4Generation opt-125mfacebook/opt-125mGeneration OTel-2.0-LLM-31B-ITfarbodtavakkoli/OTel-2.0-LLM-31B-ITGeneration gpt-oss-20bopenai/gpt-oss-20bGeneration Ornith-1.5-9B-GGUFornith-ai/Ornith-1.5-9B-GGUFGeneration