DeepSeek: DeepSeek V4 Flash

deepseek/deepseek-v4-flash

DeepSeek V4 Flash is an efficiency-optimized Mixture-of-Experts model from DeepSeek with 284B total parameters and 13B activated parameters, supporting a 1M-token context window. It is designed for fast inference and...

Prompt $/1M

$0.094

Completion $/1M

$0.188

Context

1,048,576 tok

Best provider

113 tok/s · Artificial Analysis

Specs

Model id: deepseek/deepseek-v4-flash
Display name: DeepSeek: DeepSeek V4 Flash
Modality: text->text
Tokenizer: DeepSeek
Instruct type: —
Context length: 1,048,576
Max completion: —
Prompt $/1M: $0.094
Completion $/1M: $0.188
Free tier: no

Capabilities

Function callingno
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
Artificial Analysis	113	875 ms	—
DekaLLM	79	1,593 ms	2,234 ms
NextBitfastest	78	1,361 ms	1,523 ms
Alibabafastest	69	1,127 ms	1,286 ms
DeepSeek	68	840 ms	970 ms
Baidu	68	609 ms	675 ms
AtlasCloud	55	1,548 ms	1,923 ms
StreamLake	54	1,776 ms	2,168 ms
Fireworks	47	2,059 ms	2,980 ms
Novita	45	1,563 ms	1,655 ms
GMICloud	38	2,873 ms	3,547 ms
Parasail	37	748 ms	1,023 ms
WandB	34	620 ms	907 ms
Venice	30	1,928 ms	2,137 ms
Cloudflare	29	1,027 ms	1,216 ms
Morph	27	904 ms	1,391 ms
DeepInfra	24	1,009 ms	1,314 ms
Wafer	18	3,689 ms	4,558 ms
Ambient	15	934 ms	1,145 ms
SiliconFlow	14	1,615 ms	2,049 ms
AkashML	12	1,193 ms	1,398 ms
DigitalOcean	8	1,556 ms	1,945 ms
Io Net	6	2,712 ms	3,092 ms