Deepseek V4 Flash
DeepSeek V4 Flash 0731 是 DeepSeek 推出的稀疏混合专家(MoE)模型,总参数 2840 亿,激活参数 130 亿。这一重训练修订版适用于编程、推理与 Agent 工作流。
模态
参数量 (MoE)
284B总
激活 13B / token
上下文
1.05Mtokens
最大输出 944K tokens
排名与分数走势
Deepseek V4 FlashDeepseek V4 Flash High (Preview)
排名
Elo 评分
能力
工具调用 (Tools)
结构化输出
并行工具调用
视觉输入
音频输入
多服务商价格对比数据来自 OpenRouter
| 服务商 | 量化精度 | 上下文 | 输入/M | 输出/M | 缓存读取/M | 近1天可用率 |
|---|---|---|---|---|---|---|
| fp8 | 1.05M | ¥0.29 | ¥0.88 | ¥0.01 | 100.0% | |
W Wafer | nvfp4 | 1.05M | ¥0.30 | ¥2.01 | ¥0.23 | 100.0% |
| fp8 | 1.05M | ¥0.32 | ¥0.96 | ¥0.08 | 100.0% | |
| fp8 | 1.05M | ¥0.40 | ¥1.20 | ¥0.10 | 98.6% | |
| fp4 | 1.05M | ¥0.47 | ¥0.94 | ¥0.09 | 100.0% | |
| fp8 | 1.02M | ¥0.59 | ¥1.77 | ¥0.02 | 100.0% | |
| fp8 | 1.05M | ¥0.60 | ¥1.81 | ¥0.15 | 99.3% | |
| fp8 | 1.05M | ¥0.74 | ¥1.47 | ¥0.33 | 100.0% | |
| — | 1.05M | ¥0.80 | ¥1.59 | ¥0.16 | 100.0% | |
| fp8 | 1.05M | ¥0.87 | ¥1.74 | ¥0.19 | 100.0% | |
| fp32 | 1.05M | ¥0.87 | ¥1.74 | ¥0.19 | 100.0% | |
| — | 1.00M | ¥0.87 | ¥1.74 | ¥0.19 | 99.9% | |
C CoreWeave | fp8 | 262K | ¥0.87 | ¥1.81 | ¥0.33 | 100.0% |
I Inceptron | fp4 | 1.05M | ¥0.87 | ¥3.31 | ¥0.18 | 98.5% |
W Wafer | nvfp4 | 1.05M | ¥0.94 | ¥1.54 | ¥0.90 | 100.0% |
| fp8 | 1.05M | ¥0.94 | ¥1.87 | ¥0.07 | 100.0% | |
| fp8 | 1.05M | ¥0.94 | ¥1.87 | ¥0.33 | 99.9% | |
S Sail Research | fp4 | 1.05M | ¥1.00 | ¥2.01 | ¥0.28 | 99.3% |
S Sail Research | fp4 | 1.05M | ¥1.00 | ¥2.01 | ¥0.28 | 100.0% |
M Mancer 2 | fp8 | 1.05M | ¥1.00 | ¥3.35 | ¥0.33 | 99.5% |
S SiliconFlow | fp8 | 1.05M | ¥1.47 | ¥4.42 | ¥0.19 | 99.6% |
| — | 1.00M | ¥2.36 | ¥7.07 | ¥0.24 | 100.0% | |
| — | 1.00M | ¥2.50 | ¥7.51 | ¥0.25 | 100.0% | |
| fp8 | 1.05M | ¥2.80 | ¥8.39 | ¥0.18 | 99.7% | |
C Chutes | fp8 | 1.05M | ¥2.95 | ¥8.84 | ¥0.29 | 100.0% |
P PrimeIntellect | fp4 | 1.05M | ¥2.95 | ¥8.84 | ¥0.09 | 100.0% |
¥0.29
输入/M
¥0.88
输出/M
¥0.01
缓存/M
100.0%
可用率
W
Wafernvfp4¥0.30
输入/M
¥2.01
输出/M
¥0.23
缓存/M
100.0%
可用率
¥0.32
输入/M
¥0.96
输出/M
¥0.08
缓存/M
100.0%
可用率
¥0.40
输入/M
¥1.20
输出/M
¥0.10
缓存/M
98.6%
可用率
¥0.47
输入/M
¥0.94
输出/M
¥0.09
缓存/M
100.0%
可用率
¥0.59
输入/M
¥1.77
输出/M
¥0.02
缓存/M
100.0%
可用率
¥0.60
输入/M
¥1.81
输出/M
¥0.15
缓存/M
99.3%
可用率
¥0.74
输入/M
¥1.47
输出/M
¥0.33
缓存/M
100.0%
可用率
¥0.80
输入/M
¥1.59
输出/M
¥0.16
缓存/M
100.0%
可用率
¥0.87
输入/M
¥1.74
输出/M
¥0.19
缓存/M
100.0%
可用率
¥0.87
输入/M
¥1.74
输出/M
¥0.19
缓存/M
100.0%
可用率
¥0.87
输入/M
¥1.74
输出/M
¥0.19
缓存/M
99.9%
可用率
C
CoreWeavefp8¥0.87
输入/M
¥1.81
输出/M
¥0.33
缓存/M
100.0%
可用率
I
Inceptronfp4¥0.87
输入/M
¥3.31
输出/M
¥0.18
缓存/M
98.5%
可用率
W
Wafernvfp4¥0.94
输入/M
¥1.54
输出/M
¥0.90
缓存/M
100.0%
可用率
¥0.94
输入/M
¥1.87
输出/M
¥0.07
缓存/M
100.0%
可用率
¥0.94
输入/M
¥1.87
输出/M
¥0.33
缓存/M
99.9%
可用率
S
Sail Researchfp4¥1.00
输入/M
¥2.01
输出/M
¥0.28
缓存/M
99.3%
可用率
S
Sail Researchfp4¥1.00
输入/M
¥2.01
输出/M
¥0.28
缓存/M
100.0%
可用率
M
Mancer 2fp8¥1.00
输入/M
¥3.35
输出/M
¥0.33
缓存/M
99.5%
可用率
S
SiliconFlowfp8¥1.47
输入/M
¥4.42
输出/M
¥0.19
缓存/M
99.6%
可用率
¥2.36
输入/M
¥7.07
输出/M
¥0.24
缓存/M
100.0%
可用率
¥2.50
输入/M
¥7.51
输出/M
¥0.25
缓存/M
100.0%
可用率
¥2.80
输入/M
¥8.39
输出/M
¥0.18
缓存/M
99.7%
可用率
C
Chutesfp8¥2.95
输入/M
¥8.84
输出/M
¥0.29
缓存/M
100.0%
可用率
P
PrimeIntellectfp4¥2.95
输入/M
¥8.84
输出/M
¥0.09
缓存/M
100.0%
可用率
📊 Artificial Analysis 综合指数
智能指数
编程指数
Agent 指数