AI Cost Calculators
DeepSeek API Cost Calculator
Estimate DeepSeek V4 Flash and V4 Pro API costs using cache-hit input, cache-miss input, output tokens, requests, and monthly usage.
Enter Your DeepSeek API Usage
Use average token values for one request, then enter the expected number of requests in one month.
Price used per 1 million tokens
Cache hit:$0.002800
Cache miss:$0.14
Output:$0.28
Estimated DeepSeek API Cost
This estimate covers the token charges entered above.
Estimated monthly cost
Per request
$0.000274
Per day
$0.7314
Per year
$263.2896
Cache-hit input cost
36,000,000 tokens
$0.1008
Cache-miss input cost
84,000,000 tokens
$11.76
Output cost
36,000,000 tokens
$10.08
Requests: 80,000
Total input tokens: 120,000,000
Total output tokens: 36,000,000
* Important: Calculated from the values currently shown. Default usage values are examples, so change them to match your expected usage. Built-in DeepSeek rates were checked on June 18, 2026. Final charges may include taxes, discounts, retries, price changes, and other services not entered here.
DeepSeek API costs can change with the model, input size, output length, request count, and cache-hit rate. This calculator brings those values together so you can test a small launch, a normal month, and a higher-usage case before you build.
How the DeepSeek API Cost Calculator Works
Choose DeepSeek V4 Flash or V4 Pro. Enter the expected number of requests, average input tokens, average output tokens, and the share of input tokens that may be cache hits.
The calculator separates cache-hit input, cache-miss input, and output costs. It then shows the estimated cost per request, day, month, and year.
You can also turn on custom pricing when DeepSeek updates its rates or when you want to test a private rate.
What to Enter for a Useful Estimate
Use average values from a real request. Include the system message, prompt, chat history, retrieved text, and other content sent to the model.
For the cache-hit percentage, use a careful estimate. A repeated system prompt or shared document prefix may improve cache use, but DeepSeek does not guarantee a hit for every repeated request.
Test at least three cases: a small launch, a normal month, and a busy month.
Common Ways to Use This Calculator
- Estimate the monthly cost of a DeepSeek chatbot.
- Compare DeepSeek V4 Flash and V4 Pro.
- Test the possible saving from context caching.
- Estimate cost per request, day, month, and year.
- Review the effect of longer prompts or answers.
- Prepare an early DeepSeek API budget.
Simple DeepSeek API Cost Example
Imagine an AI assistant with 80,000 requests per month. Each request uses 1,500 input tokens and 450 output tokens. If 30% of the input may be a cache hit, enter those values and choose DeepSeek V4 Flash.
You can then switch to V4 Pro without changing the workload.
Pricing and Estimate Notes
Built-in prices were checked against DeepSeek's official API pricing page on June 18, 2026.
Context caching is enabled by default, but cache hits are best effort. Real cache use may differ from the percentage entered here.
Always check the official DeepSeek API pricing page before making a final budget or purchase decision.
Frequently Asked Questions
How is DeepSeek API cost calculated?
The calculator multiplies cache-miss input, cache-hit input, and output tokens by the selected DeepSeek model rates. It then shows the estimated cost per request, day, month, and year.
What is the difference between cache-hit and cache-miss input?
Cache-hit input matches a stored prompt prefix and uses a lower rate. Cache-miss input does not match the cache and uses the normal input rate.
Is DeepSeek context caching automatic?
DeepSeek says its disk-based context cache is enabled by default. Cache hits are still best effort and are not guaranteed for every request.
Can I compare DeepSeek V4 Flash and V4 Pro?
Yes. Keep the same request and token values, then switch models to compare their estimated costs.
Are the results exact?
No. They are planning estimates. Your final bill may change because of price updates, real cache-hit rates, retries, taxes, discounts, or other services.
Can I enter my own DeepSeek prices?
Yes. Turn on custom pricing and enter your own cache-hit input, cache-miss input, and output rates per million tokens.