GoTry AI
EN 中文

DeepSeek V4.1 Flash

Language DeepSeek Available Released Sep 10, 2026 Partly verified
Model ID
Compare with… →
Context
1M
384K max output
Input price
$0.30
per 1M tokens · DeepSeek API
Output price
$1.20
per 1M tokens
Epoch Capabilities Index
155
#27 of 71 listed
LMArena Text
1,477
#22 of 79 listed

Where to run it

3 offerings · prices as listed
MeterTierPrice
DeepSeek API Official Global · USD deepseek-flash
InputOff-peak hours $0.15 / 1M tokens
Peak hours $0.30 / 1M tokens
OutputOff-peak hours $0.60 / 1M tokens
Peak hours $1.20 / 1M tokens
Cache readOff-peak hours $0.003 / 1M tokens
Peak hours $0.006 / 1M tokens
DeepSeek API Official China · CNY deepseek-flash
InputOff-peak hours ¥1.00 / 1M tokens
Peak hours ¥2.00 / 1M tokens
OutputOff-peak hours ¥4.00 / 1M tokens
Peak hours ¥8.00 / 1M tokens
Cache readOff-peak hours ¥0.020 / 1M tokens
Peak hours ¥0.040 / 1M tokens
Volcengine Ark Official cn-beijing · CNY deepseek-v4-1-flash
InputOff-peak hours ¥1.00 / 1M tokens
Peak hours ¥2.00 / 1M tokens
OutputOff-peak hours ¥4.00 / 1M tokens
Peak hours ¥8.00 / 1M tokens
Cache readOff-peak hours ¥0.020 / 1M tokens
Peak hours ¥0.040 / 1M tokens

Changes

All changes →
  1. Now priced on Volcengine Ark, cn-beijing Input (Off-peak hours) ¥1.00 · Cache read (Off-peak hours) ¥0.020 · Output (Off-peak hours) ¥4.00 · Input (Peak hours) ¥2.00 · +2 more
  2. 3 new prices on DeepSeek API Cache read (Peak hours) $0.006 · Input (Peak hours) $0.30 · Output (Peak hours) $1.20

Vendor-reported scores

Published by DeepSeek, under the conditions it states
BenchmarkSettingScore
GPQA Diamondpass@1, max reasoning effort90.9%
Humanity's Last Exampass@1, no tools, max reasoning effort36.8%
pass@1, no tools, text-only subset, max reasoning effort39.1%
pass@1, with tools, max reasoning effort63.9%
Terminal-Bench 2.1pass@1, DeepSeek Harness Minimal mode, max reasoning effort90.6%

Cited scores

From third-party leaderboards, not measured by us
  • Epoch Capabilities Index 155

    #27 of 71 listed models · CI 148.5–157.5

    Epoch AI, 'Capabilities & benchmarking'. Published online at epoch.ai. Retrieved from 'https://epoch.ai/benchmarks'. Licensed CC BY 4.0. · License · as of Sep 26, 2026 · changes we made

  • LMArena Text 1,477

    #22 of 79 listed models · max effort · CI 1,469–1,484 · 6,630 votes

    LMArena (Arena), leaderboard-dataset on Hugging Face, licensed CC BY 4.0 · License · as of Sep 25, 2026 · changes we made

  • LMArena Text · Chinese 1,525

    #21 of 79 listed models · max effort · CI 1,499–1,552 · 512 votes

    LMArena (Arena), leaderboard-dataset on Hugging Face, licensed CC BY 4.0 · License · as of Sep 25, 2026 · changes we made

  • LMArena Text · Coding 1,532

    #12 of 79 listed models · max effort · CI 1,519–1,546 · 1,938 votes

    LMArena (Arena), leaderboard-dataset on Hugging Face, licensed CC BY 4.0 · License · as of Sep 25, 2026 · changes we made