fireworks/deepseek-v4-flash

Common Name: DeepSeek V4 Flash

Fireworks
-50%On SaleSupportedTool InvocationSupportedReasoning
CompareTry in Chat

DeepSeek's cost-efficient hybrid-thinking model in the V4 family.

Specifications

Context
1000K
Maximum Output
384K
Inputtext
Outputtext

Performance (7-day Average)

Collecting…
Collecting…
Collecting…

Pricing

Input$0.07/MTokens
Cached Input$0.014/MTokens
Output$0.14/MTokens

Availability Trend (24h)

Performance Metrics (24h)

Similar Models

$0.075/$0.30/M-50%
ctx131Kmax33Kavailtps
InOutCap

OpenAI's open-weight 120B model for production, agentic tasks, and high-reasoning use cases.

$0.035/$0.15/M-50%
ctx131Kmax33Kavailtps
InOutCap

OpenAI's open-weight 20B model for lower-latency reasoning and specialized use cases.

$1.50/$7.50/M-50%
ctx1.0Mmax131Kavailtps
InOutCap

Kimi's flagship multimodal reasoning model for long-context knowledge work and agentic workflows.

$0.70/$2.20/M-50%
ctx1.0Mmax131Kavailtps
InOutCap

Zhipu AI's flagship reasoning model for long-context engineering and agentic tasks.