← All models

Qwen3 30B A3B

qwen

Sparse MoE Qwen model with 3B active parameters for efficient chat and reasoning

Context
131,072
Output
16,384
Release date
2025-04-28
Open weights
Yes

Availability from providers