Models·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...

$0.028Input, per million tokens
$0.056Output, per million tokens
1.05MContext window
384KLongest answer

1.05M tokens of context — about 786,432 words, on a scale that ends at 2M.

What it can do

  • Reasoning
  • Tool use
  • Open weights
  • Takes text
  • Gives text

Released 24 Apr 2026 · checked 3 Oct 2026.

About it

DeepSeek V4 Flash 0423 is a text model from DeepSeek, built for efficiency with a Mixture-of-Experts design. It has 284B total parameters, with 13B activated parameters, and supports a 1,048,576-token context window. The model accepts text and produces text, with reasoning and tool support. It has open weights, and its Hugging Face identifier is deepseek-ai/DeepSeek-V4-Flash. Its maximum output is 384,000 tokens. The listed input price is USD 0.028 per 1M tokens, and the output price is USD 0.056 per 1M tokens. The model was released on 2026-04-24. Its listed model ID is deepseek/deepseek-v4-flash.

Verdict

DeepSeek V4 Flash 0423 combines open weights, tool support and reasoning with a large context window. Its listed prices are USD 0.028 per 1M input tokens and USD 0.056 per 1M output tokens.

  • ["1,048,576-token context window", "Open weights", "Supports reasoning and tools", "Maximum output of 384,000 tokens"]
  • ["Text input and output only", "Input costs USD 0.028 per 1M tokens", "Output costs USD 0.056 per 1M tokens"]

On these boards

More from DeepSeek