Back to comparator
qwen

Qwen: Qwen3.5-35B-A3B

qwen/qwen3.5-35b-a3b

For $1, you can send approximately:
~97.7messages

How do we get this number?

One message = ~7,000 input tokens + ~7,000 output tokens
Input cost per message7,000 x $0.16/M = $0.001138
Output cost per message7,000 x $1.30/M = $0.009100
Total cost per message$0.010238
Messages for $197.68
Context window
262k
tokens
Max response
66k
tokens
Input price
$0.16
per million tokens
Output price
$1.30
per million tokens

Modalities

Input:textInput:imageInput:videoOutput:text

Description

The Qwen3.5 Series 35B-A3B is a native vision-language model designed with a hybrid architecture that integrates linear attention mechanisms and a sparse mixture-of-experts model, achieving higher inference efficiency. Its overall...