DeepSeek API pricing and cost calculator
Short answer: DeepSeek lists 2 models with verified prices, from $0.3 to $1.32 per 1M input tokens and $1.2 to $3.96 per 1M output tokens. For a 2,000-input / 500-output call, the cheapest DeepSeek model is DeepSeek Flash (V4.1) at $0.0012; the most expensive is DeepSeek V4 Pro at $0.00462 (checked 2026-11-15).
Nearest-priced models, this workload
Showing the 9 models closest in price; change any input to rank every model.
| Model | Per call | Per 1K calls |
|---|---|---|
| Mistral Small 4 | $0.0006 | $0.6000 |
| GPT-5.6 Luna | $0.0010 | $1.00 |
| GPT-5.4 nano (deprecated, retires 2027-04-01) | $0.00103 | $1.03 |
| Codestral 25.08 | $0.00105 | $1.05 |
| DeepSeek Flash (V4.1) | $0.0012 | $1.20 |
| Gemini 3.1 Flash-Lite | $0.00125 | $1.25 |
| GPT-5 mini (deprecated, retires 2026-12-11) | $0.0015 | $1.50 |
| Gemini 3.5 Flash-Lite | $0.00185 | $1.85 |
| Gemini 2.5 Flash | $0.00185 | $1.85 |
o4-mini is shut down; its last published list rate is kept for comparing old invoices, not as a model you can call today — the model page names its replacement.
DeepSeek prices
| Model | Input /1M | Cached /1M | Output /1M | Checked |
|---|---|---|---|---|
| DeepSeek V4 Pro | $1.32 | $0.044 | $3.96 | 2026-11-15 |
| DeepSeek Flash (V4.1) | $0.3 | $0.006 | $1.2 | 2026-11-15 |
Official list price vs cloud and reseller prices
Every rate on this page is the list price published on DeepSeek's own pricing page. The same models are also sold through cloud platforms (Microsoft Azure, Amazon Bedrock, Google Vertex AI) and through model marketplaces and resellers. Those channels publish their own price lists and can add a platform fee, regional pricing, committed-use discounts or a different batch and caching policy, so the amount on their invoice does not always match the list price here.
We keep cloud and reseller offers out of the table because we have not read them on the operator's own page and dated them; mixing them with verified list prices would make the comparison look more precise than it is. If you buy through a cloud or a reseller, use this page as the list-price baseline and check that channel's own page for what you will actually be billed. The methodology page explains the rule, and calculator vs marketplace vs provider page compares the three sources.
Browse all 2 DeepSeek models, cheapest first: DeepSeek models and prices
DeepSeek bills 5 more services without tokens: search API (text), called directly at $0.75 per 1,000 successful requests, fetch API, called directly at $0.75 per 1,000 successful requests, file parsing and OCR at $0.225 per 1,000 pages, text to speech at $0.45 per 10,000 characters and speech to text at $0.116 per hour of audio (checked 2026-11-18). Server-side search and web fetch that a model call triggers are no longer billed from 12:00 November 18, 2026 (Beijing time, UTC+8); only token usage is charged for them. Direct Search and Fetch API calls are billed at the per-request rates listed in the table. The dated table is on the DeepSeek models and prices page.
Questions
Which DeepSeek model is the cheapest?
DeepSeek Flash (V4.1) costs $0.3 per 1M input tokens and $1.2 per 1M output tokens, or $0.0012 for a 2,000-input / 500-output call.
How much do 1,000 calls cost on the cheapest DeepSeek model?
1,000 calls at 2,000 input / 500 output tokens each cost $1.20 on DeepSeek Flash (V4.1) at list price.
What counts as “DeepSeek API pricing” on this page?
Only the list price on DeepSeek's own pricing page: standard-tier input, cached input (where the provider lists one) and output per 1M tokens, with the check date on every row. Cloud platforms, marketplaces and resellers publish their own price lists and are not mixed in. If a rate cannot be read from the official page it is marked unverified instead of guessed.
What does “per token” mean in DeepSeek API pricing?
Rates are published per 1,000,000 tokens, so divide by 1M for one token. DeepSeek Flash (V4.1) at $0.3 per 1M input tokens works out at $1 for every 3,333,333 input tokens; its $1.2 output rate is $1 for every 833,333 output tokens. Multiply those counts by what your logs show — not by words in the prompt.
How current are the DeepSeek prices?
Every price is copied from the official DeepSeek pricing page and carries its own check date; the table below shows the date for each model. The most recent site-wide check was 2026-11-18.
Next step → Turn this into a monthly budget
Price data last checked 2026-11-18 against official sources · how we verify
Price-change alerts
One short email when a provider changes a price or a promo ends. No spam, unsubscribe any time.