Qwen: Qwen3 Coder 480B A35B

qwen/qwen3-coder

Qwen3-Coder-480B-A35B-Instruct is a Mixture-of-Experts (MoE) code generation model developed by the Qwen team. It is optimized for agentic coding tasks such as function calling, tool use, and long-context reasoning over...

Prompt $/1M

$0.300

Completion $/1M

$1.000

Context

1,048,576 tok

Best provider

93 tok/s · Venice

Specs

Model id: qwen/qwen3-coder
Display name: Qwen: Qwen3 Coder 480B A35B
Modality: text->text
Tokenizer: Qwen3
Instruct type: —
Context length: 1,048,576
Max completion: 65,536
Prompt $/1M: $0.300
Completion $/1M: $1.000
Free tier: no

Capabilities

Function callingyes
JSON modeno
System promptno
Visionno
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
Venicefastest	93	489 ms	526 ms
DeepInfra	87	246 ms	289 ms
WandB	62	249 ms	326 ms
AtlasCloud	56	877 ms	1,017 ms
Alibaba	41	1,417 ms	2,442 ms
Novita	27	1,951 ms	2,677 ms
Together	24	1,740 ms	1,878 ms
Google	10	11,120 ms	11,771 ms
SiliconFlow	9	3,021 ms	3,917 ms