DeepSeek API pricing and limits

www.deepseek.com · LLM APIs

SEPT 2026 audit

Hosted inference for the DeepSeek model family, priced per million tokens with peak and off-peak rates.

No affiliate links, sponsored placements or paid rankings appear on this site. Ordering follows the sourced data and the stated criteria.

Entry paid planNo entry plan — DeepSeek publishes one unified pricing standard for the API and states there are no tiered plans. Stated in the API FAQ answer on whether higher-rate-limit plans exist.

CategoryLLM APIs

Reviewed by Palash BagchiUpdated September 202611 criteria comparedSources last checked 3 October 2026

DeepSeek API pricing at a glance

Entry paid plan: No entry plan — DeepSeek publishes one unified pricing standard for the API and states there are no tiered plans source ↗
Pricing model: Per million tokens, with every rate doubling at peak: off-peak is half price, and peak is 01:00-04:00 and 06:00-10:00 UTC on weekdays source ↗

First line of each sourced value. The full text is in DeepSeek API pricing below.

DeepSeek API prices by model

API prices per modelPrices in USD per 1M tokens as published on DeepSeek's Models & Pricing page: cache-miss input and output at the peak rate. Notes gives the off-peak rate and the cache-hit input price. Off-peak rates are half the peak rates; peak hours are 01:00-04:00 and 06:00-10:00 UTC, Monday to Friday, excluding Chinese public holidays.

Model	Input (per 1M tokens)	Output (per 1M tokens)	Notes
deepseek-flash	$0.30	$1.20	Peak rate; off-peak $0.15 in / $0.60 out. Cache-hit input $0.006 ($0.003 off-peak). Served by DeepSeek-V4.1-Flash.
deepseek-v4-pro	$1.32	$3.96	Peak rate; off-peak $0.66 in / $1.98 out. Cache-hit input $0.044 ($0.022 off-peak). Model version DeepSeek-V4-Pro-0813.

Models & Pricing | DeepSeek API Docs · read 25 Sept 2026 · every row quoted from this page

This page is not published yet.

— rate_limits was verified 41 days ago; the limit for this fact is 30.
— minimum_paid_price was verified 34 days ago; the limit for this fact is 30.
— pricing_model was verified 34 days ago; the limit for this fact is 30.
— flagship_input_price was verified 34 days ago; the limit for this fact is 30.
— flagship_output_price was verified 34 days ago; the limit for this fact is 30.
— budget_input_price was verified 34 days ago; the limit for this fact is 30.
— context_window was verified 34 days ago; the limit for this fact is 30.
— max_output_tokens was verified 34 days ago; the limit for this fact is 30.
— prompt_caching was verified 34 days ago; the limit for this fact is 30.
— concurrency_limit was verified 34 days ago; the limit for this fact is 30.
— gdpr_data_residency was verified 34 days ago; the limit for this fact is 30.

01.Evaluation guide

Who it suits

Batch-shaped work that can run outside weekday mornings UTC, where off-peak halves an already low rate and a cache hit costs about 3% of a cache miss.

Consider something else if…

Capacity is expressed as concurrent requests rather than tokens per minute, which is a different thing to plan against, and the peak window lands on European working hours.

Editorial · Palash Bagchi · approved

Machine-derived signals

Not sourced facts — counts and matches read by a script, not a person. See each line for where it comes from.

DeepSeek's own site (deepseek.com) self-detects as: hosting — Chinese hosting, Server Location, Hong Kong hosting, German hosting, Cloud Hosting · CDN — Edge Delivery Network · DNS — Enterprise DNS (category-level via BuiltWith, not a named vendor)

Each value below links to the page it was read from, with the sentence it came from. Criteria DeepSeek does not publish are not listed.

DeepSeek API pricing

2 criteria

Sources (1) →

Are there plans with higher rate limits? — DeepSeek Help Center (API) ↗
“There is a unified pricing standard and no tiered plans.”
Read 2026-09-13 · official docs · Report a correction

Pricing modelPer million tokens, with every rate doubling at peak: off-peak is half price, and peak is 01:00-04:00 and 06:00-10:00 UTC on weekdays.

Sources (1) →

Models & Pricing | DeepSeek API Docs ↗
“The prices listed below are in units of per 1M tokens. [...] Off-peak rates are half of the peak rates. Peak hours are 01:00 - 04:00 and 06:00 - 10:00 UTC, Monday through Friday (all other hours are off-peak).”
Read 2026-09-13 · official pricing · Report a correction

Inference

7 criteria

Input price, top model$1.32 per million input tokens for deepseek-v4-pro at peak on a cache miss, halving to $0.66 off-peak; a cache hit is $0.044 peak and $0.022 off-peak.

Sources (1) →

Models & Pricing | DeepSeek API Docs ↗
“1M INPUT TOKENS (CACHE MISS) OFF-PEAK $0.15 $0.66 PEAK $0.3 $1.32 [...] 1M INPUT TOKENS (CACHE HIT) OFF-PEAK $0.003 $0.022 PEAK $0.006 $0.044”
Read 2026-09-13 · official pricing · Report a correction

Output price, top model$3.96 per million output tokens for deepseek-v4-pro at peak, $1.98 off-peak.

Sources (1) →

Models & Pricing | DeepSeek API Docs ↗
“1M OUTPUT TOKENS OFF-PEAK $0.6 $1.98 PEAK $1.2 $3.96”
Read 2026-09-13 · official pricing · Report a correction

Input price, cheapest model$0.3 per million input tokens for deepseek-flash at peak on a cache miss, $0.15 off-peak; output is $1.2 peak and $0.6 off-peak.

Sources (1) →

Models & Pricing | DeepSeek API Docs ↗
“MODEL deepseek-flash(1) deepseek-v4-pro(2) [...] 1M INPUT TOKENS (CACHE MISS) OFF-PEAK $0.15 $0.66 PEAK $0.3 $1.32 [...] 1M OUTPUT TOKENS OFF-PEAK $0.6 $1.98 PEAK $1.2 $3.96”
Read 2026-09-13 · official pricing · Report a correction

Context window1M tokens on both models in the current line, deepseek-flash and deepseek-v4-pro.

Sources (1) →

Models & Pricing | DeepSeek API Docs ↗
“CONTEXT LENGTH 1M”
Read 2026-09-13 · official pricing · Report a correction

Max output tokens384K tokens maximum.

Sources (1) →

Models & Pricing | DeepSeek API Docs ↗
“MAX OUTPUT MAXIMUM: 384K”
Read 2026-09-13 · official pricing · Report a correction

Prompt cachingSupported

Sources (1) →

Models & Pricing | DeepSeek API Docs ↗
“1M INPUT TOKENS (CACHE HIT) OFF-PEAK $0.003 $0.022 PEAK $0.006 $0.044”
Read 2026-09-13 · official pricing · Report a correction

Rate limitsConcurrency rather than requests per minute: 500 concurrent requests on deepseek-v4-pro and 2,500 on deepseek-v4-flash, per account rather than per key. Exceeding it returns HTTP 429, and expansion is free on request.

Sources (1) →

Rate Limit & Isolation — DeepSeek API Docs ↗
“Concurrency Limit | 500 | 2500 | 2500 [...] Concurrency limits are calculated at the account level, regardless of which API Key is used [...] when the concurrency limit is exceeded, you will receive an HTTP 429 error code [...] There is no additional cost for capacity expansion.”
Read 2026-09-06 · official docs · Report a correction

Queues

1 criteria

Concurrency2,500 concurrent requests on deepseek-flash and 500 on deepseek-v4-pro.

Sources (1) →

Models & Pricing | DeepSeek API Docs ↗
“Concurrency Limit(4) 2500 500”
Read 2026-09-13 · official pricing · Report a correction

Compliance

1 criteria

GDPR / data residencyPersonal data is collected, processed and stored in the People's Republic of China; Prighter is the appointed EU and UK privacy representative.

Sources (1) →

DeepSeek Privacy Policy ↗
“we directly collect, process and store your Personal Data in People's Republic of China [...] have therefore appointed Prighter Group with its local partners as our privacy representative”
Read 2026-09-13 · official site · Report a correction

Confirm by email; unsubscribe from any issue. Your address goes to Kit and nowhere else — what we do with it.

DeepSeek in the news

Headlines from other publications, not checked by us · collected 3 Oct 2026

Selected by a search on this company, refreshed weekly. Nobody here has verified these claims — where something material is confirmed, it is written into the page above with its source, the way every other fact on this site is.

DeepSeek API pricing questions

How much does DeepSeek cost?

No entry plan — DeepSeek publishes one unified pricing standard for the API and states there are no tiered plans. Stated in the API FAQ answer on whether higher-rate-limit plans exist.

Are there plans with higher rate limits? — DeepSeek Help Center (API) · read 13 Sept 2026

How does DeepSeek charge?

Per million tokens, with every rate doubling at peak: off-peak is half price, and peak is 01:00-04:00 and 06:00-10:00 UTC on weekdays.

Models & Pricing | DeepSeek API Docs · read 13 Sept 2026

DeepSeek API pricing and limits

DeepSeek API pricing at a glance

DeepSeek API prices by model

Who it suits

Consider something else if…

DeepSeek API pricing

Inference

Queues

Compliance

DeepSeek in the news

DeepSeek API pricing questions

How much does DeepSeek cost?

How does DeepSeek charge?

Comments