← Back to Directory

DeepSeek: DeepSeek V4 Flash Latest

~DEEPSEEK Developer Architecture Profile

OpenRouter ↗HuggingFace ↗
Intelligence (ELO)1414Chatbot Arena Verified
Max Context Limit1,048,576Max Out: 943,718 tokens
Prompt CachingStandard$0.0051 / 1M cached
Standard API / 1M$1.29Blended Standard Rate

Model Capabilities & Contract SLAs

  • Drafting
  • Classification
  • Coding & Logic
  • Fictional
This model always redirects to the latest model in the DeepSeek V4 Flash family.

Granular Pricing Matrix

Input Tokens (Standard Prompt)$0.0051 / 1M
⚡ Cached Input (Prompt Cache Read)% OFF$0.0051 / 1M
Output Tokens (Completion)$1.28 / 1M

Pricing data via OpenRouter. Sync: 10/2/2026

Evaluate Competitors

VS Engine MatchupDeepSeek: DeepSeek V4 Flash Latest vs Perceptron: Perceptron Mk1.5VS Engine MatchupDeepSeek: DeepSeek V4 Flash Latest vs Sakana: Sakana NamazuVS Engine MatchupDeepSeek: DeepSeek V4 Flash Latest vs Mistral: Ministral 3 8B 2512 (batch)VS Engine MatchupDeepSeek: DeepSeek V4 Flash Latest vs Mistral: Mistral Large 3 2512VS Engine MatchupDeepSeek: DeepSeek V4 Flash Latest vs MiniMax: MiniMax M3VS Engine MatchupDeepSeek: DeepSeek V4 Flash Latest vs SpaceXAI: Grok Build 0.1