BroadRouter

DeepSeek V4 Pro API pricing

Updated

Among the 11 providers we checked (prices read 2026-11-14 to 2026-11-15), the lowest blended standard price for DeepSeek V4 Pro is GMI Cloud (promotional) at USD 0.957 input and USD 1.914 output per 1M tokens. The median across 11 standard prices is USD 1.65 / USD 3.301. DeepSeek's own API charges USD 1.32 / USD 3.96 at peak hours for the version it serves.

DeepSeek · slug deepseek-v4-pro · up to 1,048,576 tokens of context · 11 providers with a standard price. Every price is read from the provider's own pricing page and linked in its row, checked 2026-11-15. No aggregator feed. Download all prices (every provider and model, with sources, CSV or JSON). How it compares with the market: the LLM API price index. Archived list prices since 2026-05-01 (16 archived price points, 2 list-price changes): LLM API price history.

What does DeepSeek charge for DeepSeek V4 Pro?

Input
USD 1.32
Cached input
USD 0.044
Output
USD 3.96
Blended 3:1
USD 1.98

USD per 1M tokens on DeepSeek, model deepseek-v4-pro, standard tier, peak hours; off-peak USD 0.66 input, USD 0.022 cached and USD 1.98 output. Serving version: DeepSeek's API serves DeepSeek V4 Pro 0813 under this name, so its price is shown here for reference and is not ranked against hosts of DeepSeek V4 Pro. Source, checked 2026-11-14.

Is a host cheaper than DeepSeek? Yes. GMI Cloud charges USD 0.957 input and USD 1.914 output per 1M tokens, 40% below DeepSeek's own USD 1.32 / USD 3.96 at peak hours on the 3:1 blend (a promotional price). DeepSeek's row is for the version its API serves under this name (DeepSeek V4 Pro's 0813 snapshot), so it can differ slightly from what the hosts serve. Off-peak, DeepSeek charges USD 0.66 input, USD 0.022 cached and USD 1.98 output.

DeepSeek V4 Pro price per provider

USD per 1M tokens. Standard prices are ranked on a blend of 3 input tokens to 1 output token, cheapest first; the dearest standard price is 2.5× the cheapest. Free rows, the maker's serving version and other service tiers follow, tagged and unranked.

DeepSeek V4 Pro API prices by provider
#ProviderInputCached inputOutputBlended 3:1ContextPrecisionSource
1GMI Cloud cheapest promo
deepseek-ai/DeepSeek-V4-Pro
Notes and source quote

Promotional price, 45% off list (list 1.74 in / 3.48 out / 0.145 cached).

“Price (per 1M tokens) 45% off Input $0.957 $1.74 Output $1.914 $3.48 Cache Read $0.07975 $0.145 Cache Write -”
USD 0.957USD 0.07975USD 1.914USD 1.1961,048,576FP8GMI Cloud page
checked 2026-11-15
re-checked 2026-11-16
2DeepInfra
deepseek-ai/DeepSeek-V4-Pro
Notes and source quote
“$1.30 in $2.60 out $0.10 cached / 1M tokens”
USD 1.30USD 0.10USD 2.60USD 1.6251,048,576FP8DeepInfra page
checked 2026-11-14
re-checked 2026-11-16
3SiliconFlow
Notes and source quote

Context shown as 1049K.

“DeepSeek-V4-Pro 1049K $1.50162 $0.135 $3.135”
USD 1.50162USD 0.135USD 3.135USD 1.911,048,576not statedSiliconFlow page
checked 2026-11-14
re-checked 2026-11-16
4Featherless
deepseek-ai/DeepSeek-V4-Pro
Notes and source quote

Developer plan (per-token billing, from 50 USD a month in credits).

“PRICING: INPUT $1.6 / CACHED $0.2 / OUTPUT $3.2”
USD 1.60USD 0.20USD 3.20USD 2262,144FP8Featherless page
checked 2026-11-15
re-checked 2026-11-16
5Novita AI
deepseek/deepseek-v4-pro
Notes and source quote
“DeepSeek V4 Pro 1M $1.6 /Mt · Cache Read $0.135 /Mt $3.2 /Mt”
USD 1.60USD 0.135USD 3.20USD 21,048,576not statedNovita AI page
checked 2026-11-14
re-checked 2026-11-16
6Venice
deepseek-v4-pro
Notes and source quote
“| DeepSeek V4 Pro | `deepseek-v4-pro` | $1.65 | $3.301 | $0.33 | — | Private |”
USD 1.65USD 0.33USD 3.301USD 2.0631,000,000not statedVenice page
checked 2026-11-15
re-checked 2026-11-16
7Baidu
deepseek-v4-pro
Notes and source quote
“DeepSeek-V4-Pro DeepSeek-V4-Pro Inference Service Input 1.69 $/M tokens Cache Hit 0.14 $/M tokens Output 3.38 $/M tokens”
USD 1.69USD 0.14USD 3.38USD 2.1121,048,576not statedBaidu page
checked 2026-11-15
re-checked 2026-11-16
8DigitalOcean
Notes and source quote

DigitalOcean-hosted.

“DeepSeek V4 Pro — Processing modePrompt lengthInputOutputCache read: Standard All prompts$1.74$3.48$0.348”
USD 1.74USD 0.348USD 3.48USD 2.175not statednot statedDigitalOcean page
checked 2026-11-14
re-checked 2026-11-16
9Microsoft Azure
Notes and source quote

Foundry Models serverless (pay-as-you-go), Global deployment. Also sold as "FW DeepSeek V4 Pro" (Fireworks on Azure, Data Zone): input $1.925, cached $0.165, output $3.828.

“DeepSeek-V4 Pro Global $1.74 $0.145 $3.48”
USD 1.74USD 0.145USD 3.48USD 2.175not statednot statedMicrosoft Azure page
checked 2026-11-14
re-checked 2026-11-16
10Parasail
Notes and source quote
“DeepSeek V4 Pro $1.74 $3.48 $0.10”
USD 1.74USD 0.10USD 3.48USD 2.175not statednot statedParasail page
checked 2026-11-14
re-checked 2026-11-16
11Alibaba Cloud
deepseek-v4-pro
Notes and source quote

International (Singapore) deployment. Flat price, no peak/off-peak split. Alibaba lists it separately from deepseek-v4-pro-0813; which snapshot it serves is not stated. Cached price not listed (see the Model Studio console).

“deepseek-v4-pro context caching discount International $2.400 $4.800”
USD 2.40not publishedUSD 4.80USD 3not statednot statedAlibaba Cloud page
checked 2026-11-14
re-checked 2026-11-16
—DeepSeek maker, serving version peak price
deepseek-v4-pro
Notes and source quote

Peak-hour price (second price column of the quoted table). Off-peak is half: 0.66 input, 0.022 cached, 1.98 output. Peak hours are 01:00 - 04:00 and 06:00 - 10:00 UTC, Monday through Friday, excluding Chinese public holidays. All other hours are off-peak. API model name deepseek-v4-pro serves DeepSeek-V4-Pro-0813. Context 1M, max output 384K.

“(CACHE HIT) OFF-PEAK $0.003 $0.022 PEAK $0.006 $0.044 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”
USD 1.32USD 0.044USD 3.96USD 1.981,000,000not statedDeepSeek page
checked 2026-11-14
re-checked 2026-11-16
—DeepInfra flex / spot
deepseek-ai/DeepSeek-V4-Pro
Notes and source quote

Flex service tier (0.8× standard).

“Flex (0.8×) $1.04 $2.08 $0.08”
USD 1.04USD 0.08USD 2.08USD 1.301,048,576FP8DeepInfra page
checked 2026-11-14
re-checked 2026-11-16
—DeepInfra priority
deepseek-ai/DeepSeek-V4-Pro
Notes and source quote

Priority service tier (1.5× standard).

“Priority (1.5×) $1.95 $3.90 $0.15”
USD 1.95USD 0.15USD 3.90USD 2.4381,048,576FP8DeepInfra page
checked 2026-11-14
re-checked 2026-11-16

Report a wrong price or number (email to info@broadnet.ai, with this page's link filled in).

What does DeepSeek V4 Pro cost for a real workload?

Monthly cost over 30 days at each standard price, using the presets of the LLM API cost calculator. Providers without a cached price bill cached input at the input price.

  • Support chatbot: 2,000 requests a day, 1,500 input tokens (60% cached), 300 output.
  • RAG over documents: 1,000 requests a day, 8,000 input tokens (30% cached), 500 output.
  • Coding agent (one developer): 1,500 requests a day, 40,000 input tokens (85% cached), 2,000 output incl. reasoning.
Monthly cost of three workloads on DeepSeek V4 Pro, by provider
ProviderSupport chatbotRAG over documentsCoding agent (one developer)
GMI Cloud (promo)USD 73.21USD 195.23USD 552.67
DeepInfraUSD 99.00USD 264.60USD 738.00
SiliconFlowUSD 117.78USD 309.02USD 894.14
FeatherlessUSD 126.00USD 331.20USD 1,026
Novita AIUSD 122.49USD 326.52USD 926.55
VeniceUSD 136.64USD 350.48USD 1,247
BaiduUSD 129.24USD 344.70USD 974.70
DigitalOceanUSD 144.07USD 369.58USD 1,315
Microsoft AzureUSD 133.11USD 354.96USD 1,005
ParasailUSD 130.68USD 351.72USD 936.00
Alibaba CloudUSD 302.40USD 648.00USD 4,752
DeepSeek (maker, serving version, peak)USD 121.18USD 284.33USD 780.12

For the support chatbot, the cheapest price (GMI Cloud, USD 73.21 a month) saves USD 56.03 against the median (USD 129.24). For the coding agent the gap is USD 422.03 a month (43% lower).

Which DeepSeek V4 Pro provider is cheapest for your traffic?

GMI Cloud has the lowest input price (USD 0.957) and the lowest output price (USD 1.914), so it is the cheapest standard price at every mix of input and output tokens, before cache discounts.

DeepSeek V4 Pro: price per 1M tokens at 3 input to 1 output, by share of input served from cache
ProviderNo cache hits50% cached90% cached
GMI CloudUSD 1.196 cheapestUSD 0.8673 cheapestUSD 0.6041 cheapest
DeepInfraUSD 1.625USD 1.175USD 0.815
SiliconFlowUSD 1.91USD 1.397USD 0.9875
FeatherlessUSD 2USD 1.475USD 1.055
Novita AIUSD 2USD 1.451USD 1.011
VeniceUSD 2.063USD 1.568USD 1.172

USD per 1M tokens, 3 input tokens to 1 output token, each provider's cheapest standard row; 5 dearer providers are in the price table above. A provider without a cached price bills cached input at the input price. Any charge for writing to the cache, which some providers add, is not included.

10 of 11 providers publish a cached-input price below the input price: GMI Cloud (92% off), DeepInfra (92% off), SiliconFlow (91% off), Featherless (88% off), Novita AI (92% off) and 5 more. Alibaba Cloud publishes none, so cached tokens there bill at the input price.

Context windows differ: 1,048,576 tokens at GMI Cloud, DeepInfra, SiliconFlow and 2 more; 1,000,000 tokens at Venice; 262,144 tokens at Featherless. DigitalOcean, Microsoft Azure, Parasail and Alibaba Cloud do not state one. The cheapest row is not always the one that fits a long prompt.

What does DeepSeek V4 Pro cost in other languages?

The same text needs a different number of tokens in each language. With the DeepSeek V4.1 Flash tokenizer, sending 1M characters costs this much at the cheapest standard input price we found for DeepSeek V4 Pro:

ContentInput per 1M charactersCheapest provider
EnglishUSD 0.279DeepInfra
Arabic MSAUSD 0.489DeepInfra
HindiUSD 0.751DeepInfra
ChineseUSD 0.762DeepInfra
CodeUSD 0.343DeepInfra

From BroadRouter's real cost by language dataset, which covers 15 languages and content types for every model we could match to a measured tokenizer.

Published benchmark results

BroadNet did not run these benchmarks; results are as published by the DeepSeek model card and Epoch AI and may use different settings. BroadRouter is BroadNet's site.

Published benchmark results for DeepSeek V4 Pro, with sources
BenchmarkResultSetting or noteSource
MMLU-Pro (EM)87.5Think Max reasoning modeDeepSeek model card, checked 2026-11-16
GPQA Diamond (Pass@1)90.1Think Max reasoning modeDeepSeek model card, checked 2026-11-16
LiveCodeBench (Pass@1)93.5Think Max reasoning modeDeepSeek model card, checked 2026-11-16
SWE Verified (Resolved)80.6Think Max reasoning modeDeepSeek model card, checked 2026-11-16
Terminal Bench 2.0 (Acc)67.9Think Max reasoning modeDeepSeek model card, checked 2026-11-16
GPQA Diamond0.896Max reasoning setting; Epoch run deepseek-v4-pro_max, mean score on a 0 to 1 scaleEpoch AI, checked 2026-11-16

Epoch AI rows: Epoch AI, Capabilities & benchmarking, licensed CC BY 4.0, shown as the mean score on a 0 to 1 scale, rounded to three decimals (hover a value for the exact figure in Epoch's data file).

Estimate your own bill

The DeepSeek models whose cheapest standard price is closest to DeepSeek V4 Pro's (USD 1.196 blended):

DeepSeek models priced closest to DeepSeek V4 Pro
ModelCheapest inputCheapest outputMax contextProviders
DeepSeek V4 Pro 0813USD 0.85USD 2.651,048,57615
DeepSeek R1USD 0.50USD 2.18163,8404
DeepSeek R1 0528USD 0.385USD 1.625163,8405
DeepSeek R1 Distill Qwen 32BUSD 0.50USD 0.90131,0723
DeepSeek V3 0324USD 0.29USD 1.14131,0723
DeepSeek-V3.1-TerminusUSD 0.27USD 1163,8403

Priced closest from other makers: Kimi K2.7 Code (Moonshot AI, USD 1.20) · Qwen3 235B A22B Thinking 2507 FP8 (Qwen, USD 1.212) · Hy4 Preview (Tencent, USD 1.251).

Compare DeepSeek V4 Pro with

Also compared: vs DeepSeek V3.2 · vs GLM 5.3 · vs Qwen3.8 Max · vs Kimi K3 · vs GPT-6 Sol · vs DeepSeek V4 Pro 0813 · vs DeepSeek V4 Flash.

Every DeepSeek V4 Pro price comparison · All model-vs-model price comparisons

All 428 models: API prices compared · Cheapest LLM APIs right now · LLM API price index (open data) · Download all prices (CSV, JSON) · Price change log

Questions

How much does the DeepSeek V4 Pro API cost per token?

From USD 0.957 per 1M input tokens (USD 0.000000957 per token) and USD 1.914 per 1M output tokens (USD 0.000001914 per token) at the lowest standard prices we found. The median of 11 standard prices is USD 1.65 input and USD 3.301 output. Each price was read from the provider's own pricing page; the source is linked in its row.

Which provider is cheapest for DeepSeek V4 Pro?

Among the 11 providers we checked (prices read 2026-11-14 to 2026-11-15), the lowest is GMI Cloud (promotional price), ranked on a blend of 3 input tokens to 1 output token (USD 1.196 per 1M). It also has the lowest input and the lowest output price. Batch, flex, priority and fast tiers are shown in the table but not ranked.

Is DeepSeek V4 Pro cheaper from a host than from DeepSeek?

Yes. GMI Cloud charges USD 0.957 input and USD 1.914 output per 1M tokens, 40% below DeepSeek's own USD 1.32 / USD 3.96 at peak hours on the 3:1 blend (a promotional price). DeepSeek's row is for the version its API serves under this name (DeepSeek V4 Pro's 0813 snapshot), so it can differ slightly from what the hosts serve. Off-peak, DeepSeek charges USD 0.66 input, USD 0.022 cached and USD 1.98 output.

Does DeepSeek V4 Pro have cached input pricing?

10 of 11 standard prices include one. The lowest is USD 0.07975 per 1M at GMI Cloud; the median is USD 0.1375. Providers without one bill cached input at the input price. What cached input means.

What is the DeepSeek V4 Pro context window?

Up to 1,048,576 tokens. 2 standard prices come with less, down to 262,144 tokens (Featherless and Venice). Check the context column before you pick the cheapest row for long prompts. Context window calculator.

How much does a coding agent cost per month on DeepSeek V4 Pro?

For 1,500 requests a day, 40,000 input tokens (85% cached), 2,000 output incl. reasoning, over 30 days: USD 552.67 a month at GMI Cloud, USD 974.70 at the median price and USD 4,752 at Alibaba Cloud. Try your own numbers.

Is there a batch, flex or priority price for DeepSeek V4 Pro?

Yes: flex / spot from DeepInfra at USD 1.04 / USD 2.08; priority from DeepInfra at USD 1.95 / USD 3.90 per 1M input / output tokens. These tiers trade speed or availability for price (batch runs within hours, flex and spot can be slow or refused at capacity, priority and fast cost more for speed or reliability), so they are shown but not ranked.

Is DeepSeek V4 Pro free to use through an API?

No provider in our data lists a free price for DeepSeek V4 Pro as of 2026-11-15. Free tiers and credits by provider (checked 2026-11-14) lists general offers that do not name a model; all free LLM API models are on one page.

How does DeepSeek V4 Pro score on benchmarks?

As published, the DeepSeek model card reports 87.5 on MMLU-Pro (EM) and 90.1 on GPQA Diamond (Pass@1); Epoch AI measured 0.896 on GPQA Diamond in its own run, as a mean score on a 0 to 1 scale (checked 2026-11-16). BroadNet did not run these benchmarks, and sources may use different settings, so the figures are not directly comparable. Every published result we found, with its setting and source.

Read as MarkdownAsk ChatGPTAsk ClaudeAsk Perplexity