Hub/Models/qwen/qwen3.5-flash-02-23
qwentext + image + video → text

Qwen: Qwen3.5-Flash

Canonical ID: qwen/qwen3.5-flash-02-23

Compare ModelOpen on OpenRouter →

The Qwen3.5 native vision-language Flash models are built on a hybrid architecture that integrates a linear attention mechanism with a sparse mixture-of-experts model, achieving higher inference efficiency. Compared to the...

Artificial Analysis Benchmarks

Verified
No standardized Artificial Analysis benchmark scores recorded for this variant yet.

Token Pricing & Economics

Prompt Input (1M Tokens)
$0.07
Input
Completion Output (1M Tokens)
$0.26
Output
Prompt KV-Cache Read (1M Tokens)
$0.00
Cached
Context Window Capacity
976.563k tokens
Context

Supported Parameters & Features

frequency_penaltyinclude_reasoningmax_tokenspresence_penaltyreasoningresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_ktop_p