Model specification · DeepSeek

DeepSeek V4 Flash 0423

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...

SpecificationsReported values
Context window
1.0M
Maximum output
384K
Architecture
text->text
Tokenizer
DeepSeek
Knowledge cutoff
Moderated
No
Input modalities
text
Output modalities
text
Full pricingUSD where reported
PriceStandard
Input$0.089 / M
Output$0.177 / M
Benchmark indexesReported indexes
Intelligence
24.8
Coding
52
Agentic
27.9
Design Arena Elo
1190
Example workload costsStandard pricing only
WorkloadInput / outputCache shareEstimated cost
Quick check10K / 1K × 10%$0.00106
Long analysis100K / 4K × 150%$0.00603
Workday20K / 2K × 1,00030%$1.70
More from DeepSeekAll DeepSeek models
ModelContextInput / MOutput / MIntel.
DeepSeek V4 Pro 08131.0M$1.05$3.1536.3
DeepSeek V4 Flash 07311.3M$0.065$0.18034.5
DeepSeek V4 Pro 04231.0M$0.955$1.9130.9
DeepSeek V3.1 Terminus164K$0.270$1.0015.4
R164K$0.700$2.5011.4
DeepSeek V3 0324164K$0.290$1.149.7