Mmodel.kanadakurniwan
qwenBerbayar

Qwen: Qwen3 235B A22B Instruct 2507

qwen/qwen3-235b-a22b-2507

Qwen3-235B-A22B-Instruct-2507 is a multilingual, instruction-tuned mixture-of-experts language model based on the Qwen3-235B architecture, with 22B active parameters per forward pass. It is optimized for general-purpose text generation, including instruction following,...

Harga input / 1M token
$0.0875
Harga output / 1M token
$0.3500
Context length
262.144

Modalitas

Input
text
Output
text

Parameter yang didukung

frequency_penaltylogit_biaslogprobsmax_tokensmin_ppresence_penaltyrepetition_penaltyresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_ktop_logprobstop_p

Rilis

21 Juli 2025

Buka di OpenRouter ↗