Skip to main content
Watch: Stop Building Apps. Build Tools for AI Agents

Open-weight API pricing - last verified

DeepSeek V4 Pro: where to run it and what it costs

12 offers from 8 providers list DeepSeek V4 Pro at a per-token price. The lowest listed input price is $1.30 per 1M input tokens at DeepInfra (DeepSeek-V4-Pro), with output at $2.60. First-party, DeepSeek lists $1.32 input and $3.96 output per 1M tokens (peak-hour price). Prices are as listed on October 18, 2026; they change often and are not a benchmark.

1.6T total, 49B active, mixture of experts. Native context 1,000,000 tokens, per our local-models record (from the Hugging Face model card).

Every offerQuotesRun it yourselfAll open-weight models

Offers
12
Providers
8
Lowest listed input
$1.30
First-party input
$1.32

Per-token prices

Every DeepSeek V4 Pro offer, sorted by input price

Each row is one build as the provider lists it, with its own model ID: dated snapshots, Fast and turbo builds and regional rows are never merged. Sorted by listed input price in USD per 1M input tokens; every price is in that same unit. Not stated means the provider's page does not say.
12 hosted offers for DeepSeek V4 Pro, sorted by listed input price in USD per 1M input tokens
Provider and model IDInputUSD per 1M tokensOutputUSD per 1M tokensCached inputUSD per 1M tokensContext servedas the provider lists itQuantizationStatusSource
DeepInfradeepseek-ai/DeepSeek-V4-ProQuotes for this row$1.30$2.60$0.101024kInferred from an unlabelled tagNot statedNot stateddeepinfra.com
DeepInfradeepseek-ai/DeepSeek-V4-Pro-0813Quotes for this row$1.30$2.60$0.10Not statedNot statedNot stateddeepinfra.com
DeepSeek (first-party)deepseek-v4-proListed as: DeepSeek-V4-Pro-0813First-partyQuotes for this row$1.32Peak-hour price$3.96$0.0441MNot statedNot statedapi-docs.deepseek.com
Together AIdeepseek-ai/DeepSeek-V4-Pro-0813Listed as: DeepSeek V4 Pro 0813Quotes for this row$1.32$3.96$0.131,048,576NVFP4Serverless Inferencetogether.ai
Basetendeepseek-ai/DeepSeek-V4-Pro-0813Listed as: DeepSeek V4 Pro 0813Quotes for this row$1.32$3.96$0.1321048kNot statedModel APIsbaseten.co
Novitadeepseek/deepseek-v4-pro-0813Listed as: DeepSeek V4 Pro 0813Quotes for this row$1.32$3.96$0.0441Mfp8Not statednovita.ai
Cloudflare Workers AI@cf/deepseek-ai/deepseek-v4-pro-0813Quotes for this row$1.32$3.96$0.044Not statedNot statedNot stateddevelopers.cloudflare.com
Alibaba Cloud Model Studiodeepseek-v4-pro-0813Quotes for this row$1.32Busy-hour price$3.96Not statedNot statedNot statedInternationalalibabacloud.com
Novitadeepseek/deepseek-v4-proListed as: DeepSeek V4 ProQuotes for this row$1.60$3.20$0.1351Mfp8Not statednovita.ai
Azure AI FoundryDeepSeek-V4 ProQuotes for this row$1.74$3.48$0.145Not statedNot statedGlobalazure.microsoft.com
Azure AI Foundry (Fireworks on Foundry)FW DeepSeek V4 ProQuotes for this row$1.925$3.828$0.165Not statedNot statedData Zoneazure.microsoft.com
Alibaba Cloud Model Studiodeepseek-v4-proQuotes for this row$2.40$4.80Not statedNot statedNot statedInternationalalibabacloud.com
First-party
The model's own developer selling its API.
Peak-hour price
DeepSeek's listed price during its peak hours; off-peak is half, per its page. Alibaba's busy-hour rows work the same way.
Card shows N% off
A discount tag sits beside the price on the provider's catalog card. The number is the discounted one and may be temporary.
Inferred
Our reading, not the page's words: a name-only model match, a region taken from table order, or a context figure from an unlabelled tag.
Below native
The provider serves a shorter context than the model's native window in our local-models data.
Preview
The provider lists the model as a preview, for evaluation rather than production.

Evidence

The quote behind every value

Each value is copied from the provider's own page; one quote that backs several values shows once.

DeepInfra

DeepSeek-V4-Pro

Model ID

“$0.10 cached, $1.30 in, $2.60 out / 1M”
deepinfra.com

Variant, Input, Output, Cached input

“DeepSeek-V4-Pro 1024k $1.30 / $0.10 cached $2.60”
deepinfra.com

Context served

“DeepSeek-V4-Pro 1024k”
deepinfra.com
  • Inferred from an unlabelled tag. Context inferred from an unlabelled tag: the quoted pricing row carries "1024k" with no Context label.

Not stated

  • Batch: No per-model batch price listed on the pricing page.
  • Quantization: Quantization not stated on the pages used.
  • Status: No availability tier or release-stage label stated for this offer on the pages used.

Source notes: Pricing table columns: Model, Context, $ per 1M input tokens, $ per 1M output tokens.

DeepInfra

DeepSeek-V4-Pro-0813

Model ID, Variant, Input, Output, Cached input

“$0.10 cached, $1.30 in, $2.60 out / 1M”
deepinfra.com

Not stated

  • Batch: No per-model batch price listed on the pricing page.
  • Context served: Served context length not stated on the pages used.
  • Quantization: Quantization not stated on the pages used.
  • Status: No availability tier or release-stage label stated for this offer on the pages used.

Source notes: Listed separately from DeepSeek-V4-Pro on the text-generation catalog.

DeepSeek (first-party)

DeepSeek-V4-Pro-0813

Model ID

“MODEL deepseek-flash(1) deepseek-v4-pro”
api-docs.deepseek.com

Variant

“MODEL VERSION DeepSeek-V4.1-Flash-1017 DeepSeek-V4-Pro-0813”
api-docs.deepseek.com

Input

“PEAK $0.3 $1.32”
api-docs.deepseek.com

Input column

“1M INPUT TOKENS (CACHE MISS) OFF-PEAK $0.15 $0.66 PEAK $0.3 $1.32”
api-docs.deepseek.com

Output

“PEAK $1.2 $3.96”
api-docs.deepseek.com

Output column

“1M OUTPUT TOKENS OFF-PEAK $0.6 $1.98 PEAK $1.2 $3.96”
api-docs.deepseek.com

Cached input

“PEAK $0.006 $0.044”
api-docs.deepseek.com

Cached input column

“1M INPUT TOKENS (CACHE HIT) OFF-PEAK $0.003 $0.022 PEAK $0.006 $0.044”
api-docs.deepseek.com

Context served

“CONTEXT LENGTH 1M”
api-docs.deepseek.com
  • Peak-hour price. Off-peak is half: page states "Off-peak rates are half of the peak rates".

Not stated

  • Batch: No per-model batch price listed on the pricing page.
  • Quantization: Quantization not stated on the pages used.
  • Status: No availability tier or release-stage label stated for this offer on the pages used.

Source notes: Served model version is DeepSeek-V4-Pro-0813. Peak hours per the page: 01:00-04:00 and 06:00-10:00 UTC, Monday through Friday.

Together AI

DeepSeek V4 Pro 0813

Model ID

“DeepSeek DeepSeek V4 Pro 0813 deepseek-ai/DeepSeek-V4-Pro-0813 1048576 $1.32 $0.13 $3.96 NVFP4”
docs.together.ai

Variant, Input, Output, Cached input

“DeepSeek V4 Pro 0813 $1.32 $0.13 (cached) $3.96”
together.ai

Context served

“deepseek-ai/DeepSeek-V4-Pro-0813 1048576”
docs.together.ai

Quantization

“deepseek-ai/DeepSeek-V4-Pro-0813 1048576 $1.32 $0.13 $3.96 NVFP4”
docs.together.ai

Status

“Serverless Inference Most teams start with serverless inference”
together.ai

Not stated

  • Batch: No per-model batch price listed on the pricing page.

Source notes: The 0813 snapshot, not the original April V4 Pro release.

Baseten

DeepSeek V4 Pro 0813

Model ID, Context served

“DeepSeek V4 Pro 0813 deepseek-ai/DeepSeek-V4-Pro-0813 1048k 262k”
docs.baseten.co

Variant, Input, Output, Cached input

“DeepSeek V4 Pro 0813 DeepSeek V4 Pro 0813 $1.32 $1.32 $0.132 $0.132 $3.96”
baseten.co

Input column, Output column, Cached input column

“Price per 1M tokens Model Input Cache Input Output”
baseten.co

Status

“Model APIs Instant access to pre-optimized models running on the Baseten Inference Stack.”
baseten.co

Not stated

  • Batch: No per-model batch price listed on the pricing page.
  • Quantization: Quantization not stated on the pages used.

Source notes: The pricing page repeats each cell twice (responsive layout); the docs slug table supplies the model ID and context.

Novita

DeepSeek V4 Pro 0813

Model ID

“DeepSeek V4 Pro 0813 deepseek/deepseek-v4-pro-0813”
novita.ai

Variant, Input, Output, Cached input

“DeepSeek V4 Pro 0813 1M $1.32 /Mt· Cache Read$0.044 /Mt $3.96 /Mt”
novita.ai

Context served

“DeepSeek V4 Pro 0813 1M $1.32”
novita.ai

Quantization

“Provider DeepSeek Quantization fp8”
novita.ai

Not stated

  • Batch: No per-model batch price listed; the pricing page states batch inference is at an introductory 50% discount for supported models.
  • Status: No availability tier or release-stage label stated for this offer on the pages used.

Source notes: Novita also lists a separate DeepSeek V4 Pro (non-0813) row; recorded as its own offer.

Cloudflare Workers AI

deepseek-v4-pro-0813

Model ID, Variant

“@cf/deepseek-ai/deepseek-v4-pro-0813 $1.320 per M input tokens”
developers.cloudflare.com

Input, Output, Cached input

“@cf/deepseek-ai/deepseek-v4-pro-0813 $1.320 per M input tokens $0.044 per M cached input tokens $3.960 per M output tokens”
developers.cloudflare.com

Not stated

  • Batch: No per-model batch price listed on the pricing page.
  • Context served: Served context length not stated on the pages used.
  • Quantization: Quantization not stated on the pages used.
  • Status: No availability tier or release-stage label stated for this offer on the pages used.

Source notes: Workers AI bills in Neurons at $0.011 per 1,000 Neurons; the per-token figures are the page's own token-price column.

Alibaba Cloud Model Studio

deepseek-v4-pro-0813

Model ID, Output

“deepseek-v4-pro-0813 > context caching discount International Busy hours: $1.32 Idle hours: $0.66 Busy hours: $3.96 Idle hours: $1.98”
alibabacloud.com

Variant, Status

“deepseek-v4-pro-0813 > context caching discount International Busy hours: $1.32”
alibabacloud.com

Input

“deepseek-v4-pro-0813 > context caching discount International Busy hours: $1.32 Idle hours: $0.66”
alibabacloud.com
  • Busy-hour price. Idle hours $0.66.

Not stated

  • Cached input: Row links to a context caching discount but lists no cached-input price.
  • Batch: No per-model batch price listed on the pricing page.
  • Context served: Served context length not stated on the pages used.
  • Quantization: Quantization not stated on the pages used.

Novita

DeepSeek V4 Pro

Model ID

“DeepSeek V4 Pro deepseek/deepseek-v4-pro”
novita.ai

Variant, Input, Output, Cached input

“DeepSeek V4 Pro 1M $1.6 /Mt· Cache Read$0.135 /Mt $3.2 /Mt”
novita.ai

Context served

“DeepSeek V4 Pro 1M $1.6”
novita.ai

Quantization

“Provider DeepSeek Quantization fp8”
novita.ai

Not stated

  • Batch: No per-model batch price listed; the pricing page states batch inference is at an introductory 50% discount for supported models.
  • Status: No availability tier or release-stage label stated for this offer on the pages used.

Azure AI Foundry

DeepSeek-V4 Pro

Model ID, Variant, Input, Output, Cached input, Status

“DeepSeek-V4 Pro Global $1.74 $0.145 $3.48”
azure.microsoft.com

Input column, Output column, Cached input column

“Models Input (Per 1M tokens) Cached Input (Per 1M tokens) Output (Per 1M tokens)”
azure.microsoft.com

Not stated

  • Batch: No per-model batch price listed on the pricing page.
  • Context served: Served context length not stated on the pages used.
  • Quantization: Quantization not stated on the pages used.

Source notes: The row name is a display name, not an API deployment string; providerModelId holds that display name. The page does not say which DeepSeek V4 Pro snapshot is served.

Azure AI Foundry (Fireworks on Foundry)

FW DeepSeek V4 Pro

Model ID, Output

“FW DeepSeek V4 Pro Data Zone Input: $1.925/1M Tokens Cached Input: $0.165/1M Tokens Output: $3.828/1M Tokens”
azure.microsoft.com

Variant, Input, Status

“FW DeepSeek V4 Pro Data Zone Input: $1.925/1M Tokens”
azure.microsoft.com

Cached input

“FW DeepSeek V4 Pro Data Zone Input: $1.925/1M Tokens Cached Input: $0.165/1M Tokens”
azure.microsoft.com

Not stated

  • Batch: No per-model batch price listed on the pricing page.
  • Context served: Served context length not stated on the pages used.
  • Quantization: Quantization not stated on the pages used.

Source notes: Sold through Azure Foundry under the Fireworks tab; only a Data Zone price is listed for this model.

Alibaba Cloud Model Studio

deepseek-v4-pro

Model ID, Input, Output

“deepseek-v4-pro > context caching discount International $2.400 $4.800 1 million tokens”
alibabacloud.com

Variant, Status

“deepseek-v4-pro > context caching discount International $2.400 $4.800”
alibabacloud.com

Input column, Output column

“Model ID Deployment scope Input price (per 1 million tokens) Output price (per 1 million tokens)”
alibabacloud.com

Not stated

  • Cached input: Row links to a context caching discount but lists no cached-input price.
  • Batch: No per-model batch price listed on the pricing page.
  • Context served: Served context length not stated on the pages used.
  • Quantization: Quantization not stated on the pages used.

Source notes: Model Studio also lists deepseek-v4-pro-0813 (recorded separately) and Global and China-region rows at different prices.

Coverage

Not listed on the pages read

  • Fireworks AI: only the headline-model price table is covered; models outside it fall under size-based pricing and are not listed on the pages read.
  • Cerebras: the pay-as-you-go Developer tier lists only GPT OSS 120B and Qwen 3.8 27B; other models are Dedicated Inference by quote.
  • Novita says batch inference is at an introductory 50% discount for supported models, with no per-model batch prices.
  • Together AI shows a Batch API price toggle but lists no per-model batch figures.

Alternatives

Run DeepSeek V4 Pro yourself, or compare its first-party API

FAQ

How much does the DeepSeek V4 Pro API cost?

Listed input prices run from $1.30 to $2.40 per 1M input tokens across 12 offers on October 18, 2026, and output from $2.60 to $4.80 per 1M output tokens. The lowest listed input price is at DeepInfra (DeepSeek-V4-Pro). First-party: DeepSeek at $1.32 input, peak-hour price.

Which providers serve DeepSeek V4 Pro?

DeepSeek, Together AI, DeepInfra, Baseten, Novita AI, Cloudflare Workers AI, Azure AI Foundry and Alibaba Cloud Model Studio. DeepInfra, Novita AI, Azure AI Foundry and Alibaba Cloud Model Studio list more than one build (a dated snapshot, a Fast or turbo build, or a second deployment); each is its own row on this page, under the provider's own model name.

Do all providers serve DeepSeek V4 Pro's full context window?

DeepSeek V4 Pro's native context is 1,000,000 tokens in our local-models data. No provider that states a served context lists one clearly below that. Not stated by 6 offers.

Is there a batch price for DeepSeek V4 Pro?

No per-model batch price for DeepSeek V4 Pro is listed on the pages read. Novita says batch inference is at an introductory 50% discount for supported models, with no per-model batch prices. Together AI shows a Batch API price toggle but lists no per-model batch figures.

Sources

Last verified from each provider's own pages (10 sites). This is a directory of listed prices and what each provider states about the model it serves, not a benchmark. Prices change often; check the provider's page before you commit. Speed figures appear only where a provider states them, labelled Vendor-stated.

Newsletter

Know when open-weight API prices move

One email per week. New model releases, price changes and where to run them, and what they mean for what you build.

Read more about DeepSeek V4 Pro

Related

Run open-weight models locally

The self-host alternative: downloads, memory needs and commands for each model.

Open-weight model licenses

Which of these models you may use commercially, and on what terms.

LLM API cost levers

Prompt caching, batch discounts and service tiers by provider.

Free LLM API tiers

Which providers have a standing free tier or credits.

OpenAI-compatible endpoints

Base URLs and limits for switching providers without rewriting code.

LLM gateways and routers

Route one model across several of these providers with fallbacks.

Other open-weight models: DeepSeek V4.1 Flash, Qwen3.8-27B, Kimi K3, GLM-5.3, MiniMax-M3, gpt-oss-120b, gpt-oss-20b, Llama 4 Maverick, Gemma 4 31B, Mistral Large 3, NVIDIA Nemotron 3.5 Super VL 120B-A12B.

Providers on this page: DeepSeek, Together AI, DeepInfra, Baseten, Novita AI, Cloudflare Workers AI, Azure AI Foundry, Alibaba Cloud Model Studio.