Back to all models
qwen
qwenQwen: Qwen3 Coder 30B A3B Instruct
Qwen3-Coder-30B-A3B-Instruct is a 30.5B parameter Mixture-of-Experts (MoE) model with 128 experts (8 active per forward pass), designed for advanced code generation, repository-scale understanding, and agentic tool use. Built on the...
qwen/qwen3-coder-30b-a3b-instruct
Context Size
160K
Input Price
430.5 Ks/M
Output Price
1,660.5 Ks/M
Architecture
Text
Supported Parameters
frequency_penaltymax_tokenspresence_penaltyrepetition_penaltyresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_ktop_p
Details
TokenizerQwen3
Max Completion32,768 tokens
Provider Context160K tokens
ModeratedNo