qwen
Qwen: Qwen3 235B A22B Instruct 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,...
Input Cost
$0.07
per 1M tokens
Output Cost
$0.10
per 1M tokens
Context Window
262,144
tokens
Developer ID: qwen/qwen3-235b-a22b-2507
Related Models
qwen
$0.15/1M
Qwen: QwQ 32B
QwQ is the reasoning model of the Qwen series. Compared with conventional instruction-tune...
qwen
$0.09/1M
Qwen: Qwen3 30B A3B Instruct 2507
Qwen3-30B-A3B-Instruct-2507 is a 30.5B-parameter mixture-of-experts language model from Qw...
qwen
$0.12/1M
Qwen: Qwen3 VL 8B Thinking
Qwen3-VL-8B-Thinking is the reasoning-optimized variant of the Qwen3-VL-8B multimodal mode...