Does a cache hit make the whole request cheaper?

Only the input tokens actually served from cache use the cache-hit rate. The remaining input uses the cache-miss rate, and output keeps its own rate.

Which DeepSeek rate should I use for a monthly budget?

If traffic timing is unknown, start with peak rates. For scheduled work, estimate peak and off-peak traffic separately using the UTC windows above.

Is DeepSeek chat pricing the same as API pricing?

This page estimates direct API token charges. A chat product, a reseller, or a gateway can have different terms and prices.

DeepSeek API Pricing & Calculator (Oct 2026)

Estimate DeepSeek API spending using the official Flash and V4 Pro rates. Input cache hits, input cache misses, and output are billed separately. Choose the time band that applies to your requests.

Prices checked on October 14, 2026 against the official pricing page.

DeepSeek API prices per million tokens (USD)

Model	Input	Cached input	Output
deepseek-flash · Peak	$0.30	$0.006	$1.20
deepseek-flash · Off-peak	$0.15	$0.003	$0.60
deepseek-v4-pro · Peak	$1.32	$0.044	$3.96
deepseek-v4-pro · Off-peak	$0.66	$0.022	$1.98

Official DeepSeek API pricing

Estimate your DeepSeek API cost

Model

Input tokens (millions)

Output tokens (millions)

Estimated cost$5.40

Uses official list prices for input and output tokens only. Cache, long context, tools, regional uplifts, taxes, and discounts are excluded.

Compare models with cache-aware estimates

How DeepSeek API billing works

Prices are USD per million tokens. Input means a cache miss; cached input means a cache hit. The simple calculator below uses cache misses; the cross-model calculator also accepts cache-hit percentages.
Peak hours are 01:00–04:00 and 06:00–10:00 UTC, Monday through Friday, excluding Chinese public holidays. All other hours are off-peak. Split workloads that cross time bands before adding their estimates.
deepseek-flash currently routes to DeepSeek-V4.1-Flash. deepseek-v4-pro routes to DeepSeek-V4-Pro-0813. Legacy Flash aliases do not lock an older model or price.
The published table does not list Batch API prices. An unavailable Batch estimate does not mean free usage.
Charges are deducted from granted or topped-up balances, with granted balance used first. Check the provider's usage and balance records to reconcile the estimate.

Track DeepSeek API costs before limits stop requests

Quota Monitor reads DeepSeek's official API balance, including the total, granted, and topped-up amounts. Connect your account to track the remaining balance alongside other providers.

Read the DeepSeek API monitoring guide

DeepSeek API pricing FAQ

Does a cache hit make the whole request cheaper?: Only the input tokens actually served from cache use the cache-hit rate. The remaining input uses the cache-miss rate, and output keeps its own rate.
Which DeepSeek rate should I use for a monthly budget?: If traffic timing is unknown, start with peak rates. For scheduled work, estimate peak and off-peak traffic separately using the UTC windows above.
Is DeepSeek chat pricing the same as API pricing?: This page estimates direct API token charges. A chat product, a reseller, or a gateway can have different terms and prices.

Cost optimization

Batch API Pricing: OpenAI, Claude & Gemini Costs

Compare standard and Batch API costs with a calculator using official model prices. Understand turnaround time, caching, retries, and budget assumptions.

Read article

Cost optimization