AI Index

DeepSeek-V3.1

Language DeepSeek Partly verified Released Aug 21, 2025
Model ID
Context
128K
66K max output
Input price
¥4.00
per 1M tokens · Alibaba Cloud Model Studio
Output price
¥12.00
per 1M tokens
Epoch Capabilities Index
139.9
#86 of 105 listed
LMArena Text
1,417
#84 of 149 listed

Where to run it

1 offering · prices as listed
MeterTierPrice
Alibaba Cloud Model Studio (opens in a new tab) Official Available on this channel since Aug 23, 2025 China · CNY Context: 131.07KReasoning: YesTool use: YesStructured output: —Max output: 65.54KInput modalities: TextOutput modalities: TextMax thinking budget: 32.77K deepseek-v3.1
InputStandard ¥4.00 / 1M tokens
OutputStandard ¥12.00 / 1M tokens
  1. Context window 131.07K → 128K
  2. Context window 131.07K → 128K
  3. Now priced on Alibaba Cloud Model Studio, China Input ¥4.00 · Output ¥12.00
  4. Context window 128K → 131.07K

Cited scores

From third-party leaderboards, not measured by us
  • Epoch Capabilities Index 139.9

    #86 of 105 listed models · CI 136.1–143.3

  • LMArena Text 1,417

    #84 of 149 listed models · CI 1,411–1,423 · 14,540 votes

  • LMArena Text · Chinese 1,454

    #82 of 149 listed models · CI 1,438–1,470 · 1,392 votes

  • LMArena Text · Coding 1,447

    #99 of 149 listed models · CI 1,436–1,459 · 2,595 votes

  • Epoch AI, 'Capabilities & benchmarking'. Published online at epoch.ai. Retrieved from 'https://epoch.ai/benchmarks'. Licensed CC BY 4.0. · License · as of Oct 3, 2026
  • LMArena (Arena), leaderboard-dataset on Hugging Face, licensed CC BY 4.0 · License · as of Oct 2, 2026
  • changes we made