DeepSeekDeepSeekOpen weightsReleased Sep 9, 2026

DeepSeek V4.1 Flash pricing, benchmarks and cost analysis

Intelligence#28 / 66
155.9

Epoch Capabilities Index (ECI)range 153.7 to 158.6

Speed
No speed data

Throughput not measured for this model

Cost#25 / 101
$289 / mo

In $0.300Out $1.20 per 1MCache discount 98%

Context#3 / 101
1M

Context window, tokens

Ranks are among the 101 models we track (intelligence = Epoch Capabilities Index by Epoch AI, CC BY 4.0; intelligence and speed: among those with data; cost: #1 is cheapest). Monthly cost uses the default workload of 2,000 input and 500 output tokens, 10,000 requests a day, 40% cached.

Comparison summary

On the default workload (2,000 input and 500 output tokens, 10,000 requests a day, 40% cached), DeepSeek V4.1 Flash comes to $289 a month, 67% below the $873 median of the 101 models we track. It ranks #25 of 101 on cost, where #1 is the cheapest.

It scores 155.9 on the Epoch Capabilities Index (#28 of 66). Among the 20 other models within 2 points, 3 cost less, and it costs 83% less than their median ($1,716). The cheapest stable-priced model that scores at least as high is Claude Haiku 5.5 (159.3) at $113 a month, 61% less.

Cached input is 98% cheaper than fresh input, a bigger discount than the 90% median. The weights are open, so other hosts can serve it and prices may differ from the provider's own. The batch API cuts the default workload to $41.46 a month. Its listed price changed on five or more days in the last month, usually because the cheapest host changed, so check the provider before relying on the figures above.

Technical specifications

Context window
1M tokens
Input price
$0.300 / 1M
Cached input price
$0.006 / 1M
Output price
$1.20 / 1M
Cache discount
98%
Batch API
Available
Batch input / cached / output
$0.050 / $0.004 / $0.150
Open weights
Yes
Release date
Sep 9, 2026
Price volatility
Volatile (5+ changes in 30 days)

Benchmarks

Your monthly bill

Standard

$289 / month
Input
$108.00
Cached input
$1.44
Output
$180.00

Batch

$41.46 / month
Input
$18.00
Cached input
$0.96
Output
$22.50

On the default workload the batch API costs $41.46 a month against $289 standard.

Price history

  • Oct 22, 2026Input $0.049 → $0.300â–² 512%
  • Oct 22, 2026Output $0.375 → $1.20â–² 220%
  • Oct 22, 2026Cached input $0.001 → $0.006â–² 500%
  • Oct 21, 2026Input $0.056 → $0.049â–¼ 12%
  • Oct 21, 2026Output $0.120 → $0.375â–² 213%
  • Oct 21, 2026Cached input $0.002 → $0.001â–¼ 50%
  • Oct 20, 2026Input $0.038 → $0.056â–² 47%
  • Oct 20, 2026Output $0.080 → $0.120â–² 50%
  • Oct 19, 2026Input $0.065 → $0.038â–¼ 42%
  • Oct 19, 2026Output $0.144 → $0.080â–¼ 44%
  • Oct 18, 2026Input $0.016 → $0.065â–² 306%
  • Oct 18, 2026Output $0.594 → $0.144â–¼ 76%
  • Oct 17, 2026Input $0.036 → $0.016â–¼ 56%
  • Oct 17, 2026Output $0.684 → $0.594â–¼ 13%
  • Oct 17, 2026Cached input $0.005 → $0.002â–¼ 60%
  • Oct 16, 2026Input $0.024 → $0.036â–² 50%
  • Oct 16, 2026Output $0.720 → $0.684â–¼ 5%
  • Oct 15, 2026Input $0.020 → $0.024â–² 20%
  • Oct 15, 2026Output $0.891 → $0.720â–¼ 19%
  • Oct 14, 2026Input $0.044 → $0.020â–¼ 55%
  • Oct 14, 2026Output $0.175 → $0.891â–² 409%
  • Oct 14, 2026Cached input $0.001 → $0.005â–² 400%
  • Oct 13, 2026Input $0.150 → $0.044â–¼ 71%
  • Oct 13, 2026Output $0.600 → $0.175â–¼ 71%
  • Oct 13, 2026Cached input $0.003 → $0.001â–¼ 67%
  • Oct 12, 2026Input $0.300 → $0.150â–¼ 50%
  • Oct 12, 2026Output $1.20 → $0.600â–¼ 50%
  • Oct 12, 2026Cached input $0.006 → $0.003â–¼ 50%
  • Oct 9, 2026Input $0.036 → $0.300â–² 733%
  • Oct 9, 2026Output $1.00 → $1.20â–² 20%
  • Oct 9, 2026Cached input $0.010 → $0.006â–¼ 40%
  • Oct 8, 2026Input $0.050 → $0.036â–¼ 28%
  • Oct 8, 2026Output $1.20 → $1.00â–¼ 17%
  • Oct 8, 2026Cached input $0.020 → $0.010â–¼ 50%
  • Oct 7, 2026Input $0.013 → $0.050â–² 285%
  • Oct 7, 2026Output $1.32 → $1.20â–¼ 9%
  • Oct 7, 2026Cached input $0.013 → $0.020â–² 54%
  • Oct 6, 2026Input $0.300 → $0.013â–¼ 96%
  • Oct 6, 2026Output $1.20 → $1.32â–² 10%
  • Oct 6, 2026Cached input $0.006 → $0.013â–² 117%
  • Oct 5, 2026Input $0.003 → $0.300â–² 9900%
  • Oct 5, 2026Output $2.40 → $1.20â–¼ 50%
  • Oct 5, 2026Cached input $0.003 → $0.006â–² 100%
  • Oct 4, 2026Input $0.300 → $0.003â–¼ 99%
  • Oct 4, 2026Output $1.20 → $2.40â–² 100%
  • Oct 4, 2026Cached input $0.006 → $0.003â–¼ 50%
  • Oct 2, 2026Input $0.030 → $0.300â–² 900%
  • Oct 2, 2026Output $0.500 → $1.20â–² 140%
  • Oct 2, 2026Cached input $0.010 → $0.006â–¼ 40%
  • Oct 1, 2026Input $0.020 → $0.030â–² 50%
  • Oct 1, 2026Output $0.396 → $0.500â–² 26%
  • Oct 1, 2026Cached input $0.003 → $0.010â–² 233%
  • Sep 30, 2026Input $0.300 → $0.020â–¼ 93%
  • Sep 30, 2026Output $1.20 → $0.396â–¼ 67%
  • Sep 30, 2026Cached input $0.006 → $0.003â–¼ 50%
  • Sep 29, 2026Input $0.035 → $0.300â–² 757%
  • Sep 29, 2026Output $0.290 → $1.20â–² 314%
  • Sep 29, 2026Cached input $0.001 → $0.006â–² 500%
  • Sep 27, 2026Input $0.300 → $0.035â–¼ 88%
  • Sep 27, 2026Output $1.20 → $0.290â–¼ 76%
  • Sep 27, 2026Cached input $0.006 → $0.001â–¼ 83%
  • Sep 26, 2026Input $0.150 → $0.300â–² 100%
  • Sep 26, 2026Output $0.600 → $1.20â–² 100%
  • Sep 26, 2026Cached input $0.003 → $0.006â–² 100%
  • Sep 25, 2026Input $0.140 → $0.150â–² 7%
  • Sep 25, 2026Output $0.420 → $0.600â–² 43%
  • Sep 25, 2026Cached input $0.004 → $0.003â–¼ 25%
  • Sep 24, 2026Input $0.150 → $0.140â–¼ 7%
  • Sep 24, 2026Output $0.600 → $0.420â–¼ 30%
  • Sep 24, 2026Cached input $0.003 → $0.004â–² 33%

Cheaper alternatives

Models with stable pricing and a Epoch Capabilities Index within 2 points of DeepSeek V4.1 Flash (or higher) that cost less for the default workload.