DeepSeek V4 Flash 0731
Review DeepSeek V4 Flash 0731 on key metrics including price, context length, throughput, and model features.
AuthorDeepseek
Context Length1.0M
Supports Tools
DeepSeek V4 Flash 0731 is the official release of DeepSeek V4 Flash, superseding the preview version, with substantially enhanced agentic capabilities. It is a sparse mixture-of-experts model with 13B active parameters out of 284B total, and this re-post-trained revision is suited for coding, reasoning, and agent workflows. The model natively supports a 1M-token context window and flexible reasoning effort: low for simple tasks, high for daily agent workflows, and max for complex ones.
Activity
Last 14 daysPrompt
7MCompletion
76KTotal
7MStartup
Startup
Latency (p50)0.88s
Throughput (p50)176.8 tok/s
Pricing
Input$0.03/M tokens
Output$0.09/M tokens
Cached input$0.008/M tokens
Features
Input Modalitiestext
Output Modalitiestext
Supported EndpointsChat Completions
Vision
Supports Tools