DeepSeek API pricing

As of November 14, 2026, the DeepSeek API costs $0.14 to $1.32 per million input tokens and $0.28 to $3.96 per million output tokens, across 7 DeepSeek models. DeepSeek Coder costs $0.14 input and $0.28 output; DeepSeek V3.2 costs $0.28 input and $0.4 output; DeepSeek V3 costs $0.27 input and $1.1 output; DeepSeek Flash costs $0.3 input and $1.2 output; DeepSeek V4 Flash costs $0.3 input and $1.2 output; DeepSeek R1 costs $0.55 input and $2.19 output; DeepSeek V4 Pro costs $1.32 input and $3.96 output. A typical request of 1,000 input and 500 output tokens costs $0.00028 on DeepSeek Coder and $0.0033 on DeepSeek V4 Pro.

Prices updated

DeepSeek API price per 1M tokens, every model

ModelInputCached inputOutputContextTypical request
DeepSeek Coder
deepseek-coder
$0.14$0.014$0.28128K$0.00028
DeepSeek V3.2
deepseek-v3.2
$0.28$0.028$0.4164K$0.00048
DeepSeek V3
deepseek-v3
$0.27$0.07$1.165.5K$0.00082
DeepSeek Flash
deepseek-flash
$0.3$0.006$1.21M$0.0009
DeepSeek V4 Flash
deepseek-v4-flash
$0.3$0.006$1.21M$0.0009
DeepSeek R1
deepseek-r1
$0.55$0.14$2.1965.5K$0.00164
DeepSeek V4 Pro
deepseek-v4-pro
$1.32$0.044$3.961M$0.0033

US dollars per million tokens, sorted by the cost of a typical request of 1,000 input and 500 output tokens.

Calculate the cost of your own prompt →

How much does a DeepSeek API request cost?

A typical request of 1,000 input and 500 output tokens costs $0.00028 on DeepSeek Coder, $0.00048 on DeepSeek V3.2, $0.00082 on DeepSeek V3, $0.0009 on DeepSeek Flash, $0.0009 on DeepSeek V4 Flash, $0.00164 on DeepSeek R1 and $0.0033 on DeepSeek V4 Pro. A coding agent step with 50K tokens of context and 2K output tokens costs $0.0739 on DeepSeek V4 Pro.

ModelShort chat message
300 in, 200 out
Typical request
1K in, 500 out
Long document summary
20K in, 1K out
Coding agent step
50K in, 2K out
1,000 typical requests
DeepSeek Coder$0.000098$0.00028$0.00308$0.00756$0.28
DeepSeek V3.2$0.000164$0.00048$0.006$0.0148$0.48
DeepSeek V3$0.000301$0.00082$0.0065$0.0157$0.82
DeepSeek Flash$0.00033$0.0009$0.0072$0.0174$0.90
DeepSeek V4 Flash$0.00033$0.0009$0.0072$0.0174$0.90
DeepSeek R1$0.000603$0.00164$0.0132$0.0319$1.65
DeepSeek V4 Pro$0.00119$0.0033$0.0304$0.0739$3.30

At standard rates, for the newest model of each DeepSeek family. A token is about 4 characters of English text, so 1,000 tokens is roughly 750 words. DeepSeek counts tokens with its own tokenizer, so the same text can be a different number of tokens.

How batch, prompt caching and long context change the price

Batch API. The price list has no batch API prices for DeepSeek models. See batch API pricing.

Prompt caching. All 7 priced DeepSeek models charge less for cached input, the part of a prompt that repeats across requests: typically 90% less. DeepSeek V4 Pro charges $0.044 per million cached input tokens instead of $1.32. See prompt caching pricing.

Long context. DeepSeek models accept 65.5K to 1M input tokens. The price list shows no long-context surcharge for DeepSeek: each model has one price up to its full context window. Filling a 1M-token window is still expensive, though: the input alone costs $1.32 on DeepSeek V4 Pro.

DeepSeek API price changes

Since October 22, 2026, when apicost.app started tracking, no DeepSeek model has changed its standard API price.

Dated from the day a change reached the price list. Every provider: LLM API price changes.

DeepSeek API vs OpenAI and Claude

It depends on the model. The cheapest DeepSeek model, DeepSeek Coder, costs $0.00028 per typical request; the cheapest OpenAI model, GPT-5 nano, costs $0.00025. DeepSeek V4 Pro ($0.0033) is closest in price to GPT-6.1 Terra ($0.0035): about the same.

DeepSeek modelInput / outputTypical requestClosest OpenAI modelClosest Claude model
DeepSeek Coder$0.14 / $0.28$0.00028GPT-5 nano $0.05 / $0.4 · $0.00025 · 1.1× cheaperClaude Haiku 5.5 $0.1 / $0.5 · $0.00035 · 1.3× the price
DeepSeek V3.2$0.28 / $0.4$0.00048GPT-4o mini $0.15 / $0.6 · $0.00045 · about the sameClaude Haiku 5.5 $0.1 / $0.5 · $0.00035 · 1.4× cheaper
DeepSeek V3$0.27 / $1.1$0.00082GPT-5.4 nano $0.2 / $1.25 · $0.000825 · about the sameClaude Haiku 5.5 $0.1 / $0.5 · $0.00035 · 2.3× cheaper
DeepSeek Flash$0.3 / $1.2$0.0009GPT-5.4 nano $0.2 / $1.25 · $0.000825 · about the sameClaude Haiku 5.5 $0.1 / $0.5 · $0.00035 · 2.6× cheaper
DeepSeek V4 Flash$0.3 / $1.2$0.0009GPT-5.4 nano $0.2 / $1.25 · $0.000825 · about the sameClaude Haiku 5.5 $0.1 / $0.5 · $0.00035 · 2.6× cheaper
DeepSeek R1$0.55 / $2.19$0.00164GPT-5 mini $0.25 / $2 · $0.00125 · 1.3× cheaperClaude Haiku 4.5 $1 / $5 · $0.0035 · 2.1× the price
DeepSeek V4 Pro$1.32 / $3.96$0.0033GPT-6.1 Terra $1 / $5 · $0.0035 · about the sameClaude Haiku 4.5 $1 / $5 · $0.0035 · about the same

For each DeepSeek model, the model from the other provider whose typical request costs the nearest, with its input / output price per million tokens, its typical request, and how that compares with the DeepSeek model.

Questions and answers

How much does the DeepSeek API cost?

As of November 14, 2026, the DeepSeek API costs $0.14 to $1.32 per million input tokens and $0.28 to $3.96 per million output tokens, across 7 DeepSeek models. DeepSeek Coder costs $0.14 input and $0.28 output; DeepSeek V3.2 costs $0.28 input and $0.4 output; DeepSeek V3 costs $0.27 input and $1.1 output; DeepSeek Flash costs $0.3 input and $1.2 output; DeepSeek V4 Flash costs $0.3 input and $1.2 output; DeepSeek R1 costs $0.55 input and $2.19 output; DeepSeek V4 Pro costs $1.32 input and $3.96 output. A typical request of 1,000 input and 500 output tokens costs $0.00028 on DeepSeek Coder and $0.0033 on DeepSeek V4 Pro.

Did DeepSeek raise its API prices?

No. Since October 22, 2026, when apicost.app started tracking, no DeepSeek model has changed its standard API price.

What is the cheapest DeepSeek model?

DeepSeek Coder (deepseek-coder), at $0.14 per million input tokens and $0.28 per million output tokens.

How much DeepSeek API use does $20 buy?

It depends on how much you use it. $20 of DeepSeek API usage pays for about 6,060 typical requests (1,000 input and 500 output tokens) on DeepSeek V4 Pro, or 71,428 on DeepSeek Coder. At 30 typical requests a day, a month costs $2.97 on DeepSeek V4 Pro and $0.25 on DeepSeek Coder. Long chats cost more per message than that, because every turn sends the whole conversation again as input tokens.

How much does a DeepSeek API request cost?

A typical request of 1,000 input and 500 output tokens costs $0.00028 on DeepSeek Coder, $0.00048 on DeepSeek V3.2, $0.00082 on DeepSeek V3, $0.0009 on DeepSeek Flash, $0.0009 on DeepSeek V4 Flash, $0.00164 on DeepSeek R1 and $0.0033 on DeepSeek V4 Pro. A coding agent step with 50K tokens of context and 2K output tokens costs $0.0739 on DeepSeek V4 Pro.

How much does prompt caching save on the DeepSeek API?

All 7 priced DeepSeek models charge less for cached input, the part of a prompt that repeats across requests: typically 90% less. DeepSeek V4 Pro charges $0.044 per million cached input tokens instead of $1.32.

Does long context cost more on the DeepSeek API?

No. DeepSeek models accept 65.5K to 1M input tokens. The price list shows no long-context surcharge for DeepSeek: each model has one price up to its full context window. Filling a 1M-token window is still expensive, though: the input alone costs $1.32 on DeepSeek V4 Pro.

Is the DeepSeek API cheaper than OpenAI?

It depends on the model. The cheapest DeepSeek model, DeepSeek Coder, costs $0.00028 per typical request; the cheapest OpenAI model, GPT-5 nano, costs $0.00025. DeepSeek V4 Pro ($0.0033) is closest in price to GPT-6.1 Terra ($0.0035): about the same.

Source: DeepSeek pricing, via the LiteLLM price list.

Other providers

OpenAI API pricingClaude API pricing (Anthropic)Gemini API pricing (Google)Mistral API pricingGrok API pricing (xAI)