Models/deepseek-v4-flash

deepseek-v4-flashActive · Cloud

Azure AI · Released Apr 2026 · DeepSeek License (MIT code)

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...

Context
1M
Max output
384K
Input
$0.19
Output
$0.51
Cached in
-
Latency
-

Capabilities

Core
Streaming
Function calling
JSON mode
Structured outputs
Multimodal
Vision
Image input
Audio
Image generation
Video
Advanced
Reasoning
Tool calling
MCP
Prompt caching
Batch API
Embeddings

Details

ProviderAzure AI
ReleasedApr 2026
LicenseDeepSeek License (MIT code)
InputText
OutputText
Throughput-
Availability100%

Related models