Loading...
Loading...
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 performance is comparable to that of the Qwen3.5-27B.
Quality Index
37.1
61st of 444
Top 14%
Coding Index
30.3
79th of 354
Top 22%
Price/1M
$0.69
423rd cheapest
129% above median
Top 63%
Speed
130 tok/s
Top 18%
TTFT
0.99s
Context Window
262K
61st largest
Top 25%
Input
$0.25
per 1M tokens
Output
$2.00
per 1M tokens
Blended
$0.69
per 1M tokens
Cheaper than 37% of models. Median price is $0.30/1M tokens.
Daily
$0.69
Monthly
$20.64
130
tokens/sec
Faster than 82% of models
0.99
seconds
Faster than 30% of models
16.34
seconds
Faster than 17% of models
Market Median
45 tok/s
187% faster
Median TTFT
0.42s
137% slower
Throughput/Dollar
189
tok/s per $/1M
Speed Comparison
Context Window
262K
tokens
Larger than 75% of models
Max Output
66K
tokens
25% of context
2.7M
1.3K
48-80 GB
A100 80GB