Qwen: Qwen3.5-35B-A3B

qwen/qwen3.5-35b-a3b

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...

Prompt $/1M

$0.140

Completion $/1M

$1.000

Context

262,144 tok

Best provider

209 tok/s · WandB

Specs

Model id: qwen/qwen3.5-35b-a3b
Display name: Qwen: Qwen3.5-35B-A3B
Modality: text+image+video->text
Tokenizer: Qwen3
Instruct type: —
Context length: 262,144
Max completion: 262,144
Prompt $/1M: $0.140
Completion $/1M: $1.000
Free tier: no

Capabilities

Function callingno
JSON modeno
System promptno
Visionyes
Moderatedno

Benchmark scores

Each axis is one benchmark, normalized to the best model in the field.

No benchmark data for this model.

Speed by provider

Provider	tok/s	TTFT	Total latency
WandBfastest	209	324 ms	416 ms
Artificial Analysis	167	997 ms	—
DeepInfra	138	270 ms	311 ms
AkashML	135	630 ms	725 ms
Parasail	111	491 ms	661 ms
Venice	109	756 ms	1,315 ms
Alibaba	100	725 ms	1,332 ms
Ambient	93	448 ms	625 ms
DekaLLM	79	1,165 ms	1,487 ms
AtlasCloud	65	934 ms	2,295 ms
NextBit	47	1,000 ms	2,201 ms
SiliconFlow	23	1,950 ms	3,245 ms