LIVE
Policy11/2OpenAI launches Grid Protectors Program to help protect US data center power grids
More ›Submit for Promotion
← Back to model catalog

DeepSeek

DeepSeek-V4-Pro

COpen

DeepSeek flagship MoE model (model version DeepSeek-V4-Pro-0813), supporting 1M context and 384K max output, with thinking mode enabled by default, and supporting JSON output, tool calling, Responses API, and Anthropic API-compatible format.

API Pricing

Input

$0.66

Output

$1.98

Cached input

$0.022

Peak hours

$3.96

USD / 1M tokens

※ This model bills by peak/off-peak: the input and output rates above are off-peak, and the peak rates are shown in the Peak hours tile.

※ Overseas pricing (DeepSeek official international pricing page https://api-docs.deepseek.com/quick_start/pricing/ , re-verified 2026-11-02): off-peak $0.66 input (cache hit $0.022) / $1.98 output per million tokens, half of the standard peak price ($1.32/$3.96). Peak hours are UTC Monday to Friday 01:00–04:00 and 06:00–10:00 (excluding Chinese statutory holidays); all other periods, including weekends and holidays, are off-peak. Search / Fetch APIs are billed at $0.75 per 1,000 successful requests each; server-side tool calls through DeepSeek Harness / Claude Code / Codex became billable per request from 2026-11-01 00:00 Beijing time.

Price sourceapi-docs.deepseek.com ↗cloud.tencent.com ↗· verified 2026-11-02

No price change in the last two snapshots

Price history

12 snapshots · 2026-04-28 → 2026-08-17

Current: $4.25Period low: $4.25 (2026-04-28)Period high: $4.25 (2026-04-28)First tracked: $4.25 (2026-04-28)

Specifications

Context window
1M Tokens
Max output
384K Tokens
Total parameters
1.6T
Active parameters
49B
Architecture
MoE
Input modalities
Text
License
MIT

Overall score

Total

49.3

Rank

#20

Tier

C

Data confidence

85%

Capability40
Scale35
Value73
Context76.9
Ecosystem77.5
Freshness0

Capabilities & use cases

SWE-bench_Verified 80.6%Codeforces 3206

Key capabilities

  • Thinking mode (enabled by default)
  • JSON output
  • Tool calling
  • Responses API
  • Anthropic API compatible
  • Context caching
  • Web search (Search API)

Best for

  • Complex reasoning
  • Long document analysis
  • Code generation
  • Agent workflows

Official links

Website ↗API platform ↗Pricing page ↗Open in comparator ↗

Same vendor

DeepSeek-V4.1-Flash

Similar price

GLM-4.6 $2Mistral Large 4 $2.09Nemotron 3 Ultra $2.2Doubao-Seed-2.0-Pro (Doubao 2.0 Pro) ¥16Qwen3.7-Plus (Qwen) $1.6Grok 4.3 $2.5

FAQ

What does DeepSeek-V4-Pro cost?
DeepSeek-V4-Pro is priced at $1.98 output and $0.66 input (USD per 1M tokens) Peak-hour output is $3.96.
How large is the context window of DeepSeek-V4-Pro?
DeepSeek-V4-Pro has a 1M Tokens context window with up to 384K Tokens output.
Where can I access DeepSeek-V4-Pro?
Official channels: Website and API platform.

Data is refreshed daily and cross-checked against two independent sources; please confirm pricing and specs on the vendor page.