DeepSeek-V3

Language DeepSeek Partly verified Released Dec 26, 2024

DeepSeek-V3 is a language model that powers DeepSeek App and can run locally. Its post-training incorporates verification and reflection patterns from the DeepSeek R1 series while controlling output style and length. Summary written by AI Index from DeepSeek’s own pages.

  • In Input modalities: Text Out Output modalities: Text
  • Reasoning
  • Reasoning can be off
  • Tool use

Prices

4 offerings · prices as listed
DeepSeek API (opens in a new tab) Official Global · USD deepseek-v3
Conditions/ 1M tokensInputCache readOutput
Standard $0.27$0.070$1.10

Prices from DeepSeek | Introducing DeepSeek-V3 (opens in a new tab) (checked Nov 7, 2026)

Alibaba Cloud Model Studio (opens in a new tab) Official Available on this channel since Jan 27, 2025 Global · USD Context: 65.54KReasoning: —Tool use: YesStructured output: —Max output: 8.19KInput modalities: TextOutput modalities: TextReasoning can be off: Yes deepseek-v3
Conditions/ 1M tokensInputOutput
Standard $0.287$1.147

Prices from Alibaba Cloud Model Studio model pricing - Alibaba Cloud Model Studio - Alibaba Cloud Documentation Center (opens in a new tab) (checked Sep 29, 2026)

Alibaba Cloud Model Studio (opens in a new tab) Official Available on this channel since Jan 27, 2025 China · CNY Context: 65.54KReasoning: —Tool use: YesStructured output: —Max output: 8.19KInput modalities: TextOutput modalities: TextReasoning can be off: Yes deepseek-v3
Conditions/ 1M tokensInputOutput
Standard no longer listed since Oct 11, 2026 ¥2.00¥8.00

Prices from 阿里云百炼:模型价格 (opens in a new tab) (checked Nov 9, 2026)

Alibaba Cloud Model Studio (opens in a new tab) Official Available on this channel since Jan 27, 2025 China · CNY vanchin/deepseek-v3
Conditions/ 1M tokensInputOutput
Standard ¥2.00¥8.00

Prices from 阿里云百炼:模型价格 (opens in a new tab) (checked Nov 9, 2026)

Standard: the price a request with no conditions pays (the headline price elsewhere on the site).

Details

Structured output
—
Family
deepseek
Model ID

Not stated on DeepSeek’s pages: max input, reasoning efforts, max thinking budget, built-in tools, knowledge cutoff, aliases.

Cited scores

From third-party leaderboards, not measured by us
  • Epoch Capabilities Index 131.8

    #113 of 123 listed models · CI 129.2–134.2

  • LMArena Text 1,358

    #122 of 163 listed models · CI 1,354–1,363 · 21,770 votes

  • Epoch AI · GPQA Diamond 56.5%

    #114 of 168 listed models · ±2.76

  • Epoch AI · MATH Level 5 64.9%

    #29 of 64 listed models · ±1.05

  • Epoch AI, 'Capabilities & benchmarking'. Published online at epoch.ai. Retrieved from 'https://epoch.ai/benchmarks'. Licensed CC BY 4.0. · License · as of Oct 18 – Nov 14, 2026
  • LMArena (Arena), leaderboard-dataset on Hugging Face, licensed CC BY 4.0 · License · as of Nov 8, 2026
  • changes we made
  1. Now priced on DeepSeek API Input $0.27 · Output $1.10 · Cache read $0.070
  2. Now priced on Alibaba Cloud Model Studio, China Input ¥2.00 · Output ¥8.00
  3. Context window 128K → 65.54K

Sources

DeepSeek’s official policies: data use, privacy, terms →

Some values on this page are still waiting to be checked against official pages: a filled dot is verified, a hollow one is not yet (with or without a source on file). How we verify data

DeepSeek · China · deepseek.com

Snapshots, the same family and other models from this provider