DeepSeek · Small model

DeepSeek V4 Flash pricing & cost calculator

Prices last updated:

Input

$0.14

per 1M tokens

Output

$0.28

per 1M tokens

Context window

1M

1,000,000 tokens

Open-weight model; price shown is DeepSeek's first-party API (cache-miss input).

About DeepSeek V4 Flash pricing

DeepSeek V4 Flash is DeepSeek’s small, low-cost model built for high-volume, latency-sensitive tasks. It is billed per token: $0.14 for every million tokens you send and $0.28 for every million tokens it generates, so output costs 2× as much as input.

For a typical request of 1,000 input and 500 output tokens, DeepSeek V4 Flash costs about $0.00028 — the cheapest of the 14 models we track. One million input tokens is roughly 2,000 pages of English prose.

DeepSeek’s official pricing page is the source of truth; see our guide to how AI pricing works for caching and long-context rules.

DeepSeek V4 Flash cost calculator

Per request
$0.00028
Per day
$0.28
Per month (30 days)
$8.40

Count tokens from your own text in the full calculator →

What DeepSeek V4 Flash costs for common tasks

Standard pricing, no caching or batch discount. Token sizes are typical estimates — see token cost by use case.
TaskInput tokensOutput tokensPer requestPer 1,000 requests
Short chat reply500300$0.000154$0.154
Support chatbot turn with history3,000400$0.000532$0.532
RAG answer (retrieved context)6,000600$0.001008$1.01
Summarize a 20-page document12,0001,000$0.00196$1.96
Long-document analysis100,0002,000$0.0146$14.56

Compare DeepSeek V4 Flash with…

Other DeepSeek models first.

DeepSeek V4 Flash pricing FAQ

How much does DeepSeek V4 Flash cost per token?
DeepSeek V4 Flash costs $0.14 per million input tokens and $0.28 per million output tokens through the DeepSeek API. That is $0.00 per input token and $0.00 per output token.
How much does a typical DeepSeek V4 Flash request cost?
A request with 1,000 input tokens and 500 output tokens costs about $0.00028 on DeepSeek V4 Flash, or $0.28 per thousand requests.
What is the context window of DeepSeek V4 Flash?
DeepSeek V4 Flash accepts up to 1,000,000 tokens (about 1M) of context per request, which covers the prompt, any documents you include and the conversation history.
Does DeepSeek V4 Flash have batch pricing?
DeepSeek does not list a batch discount for DeepSeek V4 Flash, so standard prices apply to all requests. Check DeepSeek's pricing page for off-peak or cached-input discounts.
How accurate is the token count for DeepSeek V4 Flash?
It is an estimate. DeepSeek does not publish a browser tokenizer for DeepSeek V4 Flash, so Tokenova counts with o200k_base and applies a DeepSeek-specific adjustment. Expect real counts within roughly 10–20% for English text; use DeepSeek's token-counting API for exact numbers.