Agent Cost Calculator
SourcesStart free

Peak-aware API cost planning

DeepSeek API cost calculator

Use your actual UTC request schedule or model an explicit peak-share scenario. The estimate separates cached input, uncached input and output.

Provider calculator

DeepSeek API cost calculator

Blend explicit peak and off-peak schedules instead of assuming a single always-on rate.

Effective Verified Source ↗ Direct API rates vary by peak schedule. Peak share is an explicit workload input; use the calculator’s blended estimate.

Workload and token usage
Retry rate increases the number of billed attempts.
For models without a listed cache rate, these are billed at normal input price.
Extra billed attempts after failures or repairs.
DeepSeek schedule mix
User-entered share of billed calls expected during the peak schedule.

Assumption: each meter is blended as peak rate × peak share plus off-peak rate × remaining share. Confirm the provider schedule and timezone for production budgeting.

Custom rate overrides
Enable this only when your contract or internal rate card differs from the selected public record.
Blank uses the record cache rate; if no cache rate is listed, normal input rate applies.
Assumptions

Rates are USD per million tokens unless a field says otherwise. This is a planning estimate, not a vendor quote. Records were verified 2026-11-02; region, contract, context-window, batch, tool and tax meters may differ.

Rate source ↗

No data is stored or uploaded.

Estimated monthly cost

$0.00

Enter assumptions to update the estimate.

Selected rate
$0.000Cost / request
$0.000Cost / task
0Billed attempts / month

Monthly token breakdown

New input$0.00
Cache read$0.00
5m cache write$0.00
1h cache write$0.00
Output$0.00

Separate service meters

DeepSeek API modality and usage

Token pricing and non-token usage are separate estimates. Choose a cited unit record below.

Modality and usage calculator

Separate modality usage

Use the actual unit meter for embeddings, OCR, transcription, TTS, voice or other non-token services.

Effective Verified Source ↗ USD per 1,000 successful Search API requests; separate from model token use.

Usage
Starter amount for requests; replace with measured usage.
Use this multiplier when each event contains multiple units.

The estimate is units ÷ unit size × unit price. Unit price is the cited USD price per selected unit size; it is not converted into token pricing.

No data is stored or uploaded.

Estimated monthly modality cost

$0.00

Enter usage to update the estimate.

Selected unit rate
0Units / month
$0.00Per listed unit size
—Verified date
—

Estimate, not a provider quote. Public list rates, regional terms, contract discounts, taxes and usage meters can change. Check the linked source and replace assumptions with your account terms before budgeting.

Pricing and method sources

Rate cards display their own verification and effective dates. Use the original provider source to confirm current billing details.

DeepSeek Models & Pricing ↗DeepSeek Search & Fetch ↗Methodology and exclusions →

How this estimate works

Apply the official peak-hour definition

DeepSeek’s listed direct-API prices differ between peak and off-peak hours. Peak hours are 01:00–04:00 and 06:00–10:00 UTC on weekdays, excluding Chinese public holidays. Weekends and all remaining hours are off-peak. Enter the estimated share of tokens incurred in peak windows; a single blended rate can otherwise hide meaningful schedule effects.

Count cache hits and tools as their own meters

Cache-hit and cache-miss inputs have different rates. Search and Fetch are separately charged per successful request. When using an agent harness, check whether its tool calls are billed through the API and whether the date-specific charging rule applies.

Continue with the RAG Cost Calculator if retrieval costs are part of your workflow.

Price record and update boundary

This page uses 2 token-rate record(s) verified on 2 November 2026. The data are a snapshot, not a live quote. Every selected record includes its provider source link, effective date, verification date and context note.

Frequently asked questions

When are DeepSeek peak hours?

The current official schedule defines weekday windows as 01:00–04:00 and 06:00–10:00 UTC, excluding Chinese public holidays. Other hours, weekends and full public holidays are off-peak.

How should I count cache-hit tokens?

Use the cache-hit token meter reported by the API. Do not use the prompt’s theoretical repeat share as a cache-hit count unless requests actually receive cache hits.

Are Search and Fetch included in the token estimate?

No. They are separate successful-request meters and are shown separately in the tool.

Can model names change?

Yes. The provider identifies the current V4.1 Flash and V4 Pro versions on its pricing page. Recheck model aliases and price dates before budgeting.

About the Author

Zunair Ali Syed is a doctoral researcher in Management Sciences at Minhaj University Lahore and a research consultant with experience in academic research and editing since 2011. He runs QuickEduHelp, with professional interests in digital learning, research methods, and evidence-based academic support.