Deepseek V4 Flash
Input, lowest
$0.14
Output, lowest
$0.28
Providers
1
Deepseek V4 Flash providers & prices
| Provider | Input | Output | Context | Last sync | |
|---|---|---|---|---|---|
D
DigitalOcean
|
$0.14 | $0.28 | 1,048,576 |
About Deepseek V4 Flash
DeepSeek-V4-Flash is a 284B parameter Mixture-of-Experts (MoE) model released in April 2026. It features a novel Hybrid Attention architecture (CSA+HCA) and establishes a new frontier in cost-effective 1M context processing, delivering reasoning performance that closely approaches V4-Pro when allocated an extended thinking budget
Related LLM models
gpt-oss-120b (batch)
3 providers — from $0.037/1M tokens
DeepSeek V4 Flash 0731
3 providers — from $0.06/1M tokens
Qwen3.5 397B A17B
3 providers — from $0.45/1M tokens
GLM 5.2
3 providers — from $0.75/1M tokens
Kimi K2.6
3 providers — from $0.75/1M tokens
DeepSeek V4 Pro 0813
3 providers — from $1.3/1M tokens