Skip to main content
OpenRouter logo

DeepSeek V3.1 API pricing on OpenRouter

Every OpenRouter DeepSeek V3.1 rate we track, compared against 6 providers serving the same model.

Input from /1M
$0.250
Output /1M
$0.950
Input price rank
#2 of 6
Last updated
November 21, 2026

The cheapest DeepSeek V3.1 input price we currently track is $0.200/1M on Liquid Inference.

The cheapest DeepSeek V3.1 output price is $0.700/1M on Liquid Inference.

OpenRouter DeepSeek V3.1 rates

Prices per 1M tokens. Batch and cached-input rates are shown where the provider publishes them.

OpenRouter DeepSeek V3.1 pricing by mode
ModeInputOutput
Standard$0.250/1M$0.950/1M

About DeepSeek V3.1

Creator
DeepSeek
Context
33K
Modalities
text
Open weights
Yes
Full DeepSeek V3.1 details and every provider →Compare API pricing for every DeepSeek model →

DeepSeek V3.1 on other providers

About OpenRouter

OpenRouter provides a unified API for accessing hundreds of AI models from multiple providers. It aggregates models from OpenAI, Anthropic, Google, Meta, Mistral, and many others, offering a single endpoint with automatic fallback routing and provider optimization.

Frequently Asked Questions

How much does DeepSeek V3.1 cost on OpenRouter?

OpenRouter serves DeepSeek V3.1 from $0.250 per 1M input tokens, across 1 tracked pricing mode. Prices are collected daily — see the table above for current input, output, and cached-input rates.

Is OpenRouter the cheapest way to run DeepSeek V3.1?

OpenRouter ranks #2 of 6 providers we track serving DeepSeek V3.1 on input price. The cheapest input price right now is on Liquid Inference. The lowest output price is $0.700/1M on Liquid Inference. Rates change often, and throughput, latency and context limits differ between providers — check the comparison above before committing.

Does OpenRouter offer batch or cached pricing for DeepSeek V3.1?

We track 1 DeepSeek V3.1 offering from OpenRouter. Batch and cached-input rates appear in the table above when the provider publishes them; blank cells mean we have no current figure for that mode.