Skip to content

Model

DeepSeek V4 Pro

Open MoE flagship with million-token context for coding and long agent runs

69 providers · 83 listingsreleased knowledge cutoff 2025-05open weightsbeta
$0.348
lowest input /M · routing.run
$0.696
lowest output /M · routing.run
1.05M
context window
1.05M
max output

Compare providers (sorted by listed input price)

59 of 69 providers publish a price for DeepSeek V4 Pro. The cheapest full-context option is CrofAI at $0.35/$0.8 per M — 14% below DeepSeek's own price. Listed prices span 6.5× for identical weights.

83 listings
Cache read /MCache write /MCapabilitiesStatus
routing.run
deepseek-v4-pro
$0.348$0.6961Mreasoningtoolsstructuredno visionApr 24, 2026
CrofAIvariant
deepseek-v4-pro-0813
$0.35$0.8$0.011Mreasoningtoolsstructuredno visionAug 22, 2026
CrofAIvariant
deepseek-v4-pro
$0.35$0.8$0.00301Mreasoningtoolsstructuredno visionApr 24, 2026
EBCloud
DeepSeek-V4-Pro
$0.429$0.8571Mreasoningtoolsstructuredno visionApr 24, 2026
TokenGo
deepseek/deepseek-v4-pro
$0.435$0.871Mreasoningtoolsstructuredno visionApr 24, 2026
Modelis
deepseek-v4-pro
$0.435$0.871Mreasoningtoolsstructuredno visionApr 24, 2026
Vivgrid
deepseek-v4-pro
$0.435$0.87$0.00361Mreasoningtoolsstructuredno visionApr 24, 2026
Auriko
deepseek-v4-pro
$0.435$0.87$0.00361Mreasoningtoolsstructuredno visionApr 24, 2026
LLM Gatewayvariant
deepseek/deepseek-v4-pro
$0.435$0.87$0.00361.05Mreasoningtoolsno structuredno visionApr 24, 2026
LLM Gatewayvariant
tencent/deepseek-v4-pro
$0.435$0.87$0.00361Mreasoningtoolsno structuredno visionApr 24, 2026
Alibaba (China)
deepseek-v4-pro
$0.435$0.87$0.00361Mreasoningtoolsstructuredno visionApr 24, 2026
Pioneer
deepseek-ai/DeepSeek-V4-Pro
$0.435$0.87$0.0036$0.4351Mreasoningtoolsstructuredno visionApr 24, 2026
Hugging Face
deepseek-ai/DeepSeek-V4-Pro
$0.435$0.87$0.00361.05Mreasoningtoolsstructuredno visionApr 24, 2026
Nvidia
deepseek-ai/deepseek-v4-pro
$0.435$0.87$0.00361.05Mreasoningtoolsstructuredno visionApr 24, 2026
DevPass (LLM Gateway)
deepseek-v4-pro
$0.435$0.87$0.00361.05Mreasoningtoolsstructuredno visionApr 24, 2026
DeepSeek
deepseek-v4-pro
$0.435$0.87$0.00361Mreasoningtoolsstructuredno visionAug 22, 2026
ZenMux
deepseek/deepseek-v4-pro
$0.435$0.87$0.00361Mreasoningtoolsstructuredno visionApr 24, 2026
OrcaRouter
deepseek/deepseek-v4-pro
$0.442$0.884$0.061Mreasoningtoolsstructuredno visionApr 24, 2026
AIHubMixvariant
deep-deepseek-v4-pro
$0.478$0.956$0.00431Mreasoningtoolsstructuredno visionApr 24, 2026
Vercel AI Gateway
deepseek/deepseek-v4-pro
$0.66$1.98$0.0221Mreasoningtoolsstructuredno visionApr 24, 2026

Best price history

Daily lowest published price across live providers. A lower line is better for the buyer.

Best input /M
$0.35 0.6%
Best output /M
$0.8 14.9%
Provider coverage
83
76 at the start · 24 snapshots
Best published input and output price historyDaily lowest published price per one million tokens. Accent is input price and violet is output price. The vertical scale starts at zero.$0$0.432$0.864Aug 21, 2026Sep 1, 2026Sep 14, 2026
Best input priceBest output priceAug 21, 2026Sep 14, 2026

Price badge

Current price badge for DeepSeek V4 Pro
![DeepSeek V4 Pro price](https://www.vaanalytics.in/badge/deepseek/deepseek-v4-pro.svg)
https://www.vaanalytics.in/badge/deepseek/deepseek-v4-pro.svg

Live SVG, regenerated on every hourly sync — paste it into a README to always show the current cheapest listed price. Updates when prices change; no build step needed on your side.

Recent changes

Changelog →
  • repricedvia openrouter · Sep 13, 2026
    • output: 384K → 393K tokens
    • input: $0.955 → $1.60
    • output: $1.91 → $3.20
    • cache_read: $0.08 → $0.135
  • contextvia kilo · Sep 13, 2026
    • context: 1.02M → 1.05M tokens
    • output: 384K → 393K tokens
  • new modelvia llmgateway-providers · Sep 13, 2026
  • repricedvia openrouter · Sep 8, 2026
    • input: $1.04 → $0.955
    • output: $2.07 → $1.91
    • cache_read: $0.086 → $0.08
  • new modelvia sensenova · Sep 7, 2026
  • repricedvia openrouter · Sep 7, 2026
    • input: $0.779 → $1.04
    • output: $1.56 → $2.07
    • cache_read: $0.065 → $0.086

Benchmarks

Scores a lab published about its own model are claims, not measurements. Rankings elsewhere on this site use independently measured scores only.

Weights

Permissivemit1.6T params

Standard open-source terms. Commercial use without additional conditions. Weights download without an access request.

licence and gating from the Hugging Face model card ↗ — the card is authoritative, this is a summary

Prices are per million tokens (USD). “—” means the provider does not publicly list a price for this model. Data from models.dev; verify with the provider before purchasing.