Skip to content

DeepSeek V4.1 Flash (batch)

DeepSeektext + image → textHugging Face OpenRouter

The cheapest provider for DeepSeek V4.1 Flash (batch) is Decart at $0.113 per 1M tokens (blended), charging $0.09 per 1M input and $0.18 per 1M output tokens. 29 providers serve it; the most expensive charges 7.0× as much. Context window: 1M tokens. Prices tracked since Oct 3, 2026.

Best price
$0.113
at Decart
Input / output
$0.09 / $0.18
per 1M tokens
Providers
29
7.0× spread
7 days
—
best price
30 days
—
best price
Context
1M
tokens

Providers

Active offers ranked by blended price. Bars are relative to the most expensive.

  • DecartCheapest
    $0.113
    $0.09 in · $0.18 out
    best
    1M ctx100% uptimefp4
  • $0.116
    $0.027 in · $0.383 out
    +3%
    1M ctx99.0% uptimefp8
  • $0.16
    $0.08 in · $0.40 out
    +42%
    1M ctx99.8% uptimefp4
  • $0.165
    $0.02 in · $0.60 out
    +47%
    1M ctx99.7% uptime
  • $0.18
    $0.04 in · $0.60 out
    +60%
    1M ctx99.8% uptime
  • $0.188
    $0.05 in · $0.60 out
    +67%
    1M ctx100% uptime
  • $0.19
    $0.12 in · $0.40 out
    +69%
    1M ctx99.6% uptime
  • $0.21
    $0.14 in · $0.42 out
    +87%
    1M ctx99.9% uptimefp8·−30% discount
  • $0.215
    $0.026 in · $0.78 out
    +91%
    1M ctx94.8% uptimefp4
  • $0.247
    $0.141 in · $0.564 out
    +119%
    1M ctx98.8% uptimefp8
  • $0.247
    $0.141 in · $0.564 out
    +119%
    1M ctx99.6% uptimefp8·−53% discount
  • $0.263
    $0.15 in · $0.60 out
    +133%
    1M ctx100% uptime
  • $0.313
    $0.20 in · $0.65 out
    +178%
    1M ctx98.9% uptimefp8
  • $0.315
    $0.18 in · $0.72 out
    +180%
    1M ctx99.8% uptime
  • $0.315
    $0.18 in · $0.72 out
    +180%
    1M ctx99.8% uptimefp8·−40% discount
  • $0.363
    $0.10 in · $1.15 out
    +222%
    1M ctx97.9% uptime
  • $0.368
    $0.21 in · $0.84 out
    +227%
    1M ctx100% uptimefp8
  • $0.368
    $0.21 in · $0.84 out
    +227%
    1M ctx99.7% uptime−30% discount
  • $0.398
    $0.20 in · $0.99 out
    +253%
    1M ctx99.9% uptimefp8
  • $0.42
    $0.24 in · $0.96 out
    +273%
    1M ctx99.5% uptimefp8·−20% discount
  • $0.525
    $0.30 in · $1.20 out
    +367%
    1M ctx99.7% uptime
  • $0.525
    $0.30 in · $1.20 out
    +367%
    1M ctx99.6% uptimefp8
  • $0.525
    $0.30 in · $1.20 out
    +367%
    1M ctx99.8% uptimefp8
  • $0.525
    $0.30 in · $1.20 out
    +367%
    1M ctx99.4% uptime
  • $0.525
    $0.30 in · $1.20 out
    +367%
    1M ctx99.7% uptime
  • $0.525
    $0.30 in · $1.20 out
    +367%
    1M ctx99.7% uptimefp8
  • $0.525
    $0.30 in · $1.20 out
    +367%
    1M ctx99.9% uptimefp8
  • $0.525
    $0.30 in · $1.20 out
    +367%
    1M ctx100% uptime
  • $0.525
    $0.30 in · $1.20 out
    +367%
    1M ctx99.8% uptimefp8
  • $0.788
    $0.45 in · $1.80 out
    +600%
    1M ctx100% uptime
Blended = (3 × input + output) / 4.

Price history

Hover to compare providers at any point in time. Click a legend item to hide it.

USD per 1M tokens · each step is a price change
Not enough history yet
Tracking since Oct 3, 2026. The chart fills in as new crawls arrive every 6 hours.

Long-context pricing

Higher rates once the prompt passes a size threshold.

Offered byPrompt sizeInputOutput
DeepSeekbase$0.15$0.60
> 0$0.15$0.60
> 0$0.15$0.60
> 0$0.30$1.20
> 0$0.15$0.60
> 0$0.30$1.20
> 0$0.15$0.60
Alibababase$0.30$1.20
> 0$0.30$1.20
> 0$0.15$0.60

Other pricing

Cache, reasoning and per-call fees. Token rates per 1M; others per unit.

Offered byCache read
Decart$0.018
Morph$0.008
Sail Research, Relace$0.01
InferenceNet$0.02
Wafer$0.048
DekaLLM$0.005
DeepInfra, Phala$0.0042
OpenInference$0.013
AtlasCloud$0.0141
StreamLake$0.00282
DeepSeek$0.003
CoreWeave, Alibaba +1$0.03
DigitalOcean, GMICloud$0.0036
Ionstream$0.0055
NextBit$0.004
Makora, Baidu +4$0.006
Novita$0.0048
BaseTen$0.007
Venice$0.0075
Fireworks (us)$0.009

Change log

Every recorded change for this model since Oct 3, 2026.

No changes yet
Tracking since Oct 3, 2026. Price changes, new providers and delistings will show up here.

About this model

DeepSeek V4.1 Flash is a sparse mixture-of-experts model from DeepSeek, and the first built on the company's Causal Encoder-Decoder (CED) architecture. It activates 8B parameters on input and 16B on...
Released
Sep 10, 2026
First tracked
Oct 3, 2026
Canonical slug
deepseek/deepseek-v4.1-flash-20260910
Weights
deepseek-ai/DeepSeek-V4.1-Flash