Home / Models / DeepSeek / DeepSeek V4 Flash 0731

DeepSeek V4 Flash 0731 pricing and routing profile

DeepSeek V4 Flash 0731 from DeepSeek lists at $0.021 per million input tokens and $0.32 per million output tokens, with a 1,310,720-token context window. Prices from OpenRouter's catalogue on 2026-09-28.

Input, per 1M tokens$0.021

Output, per 1M tokens$0.32

Cached input, per 1M$0.016

Context window1.31M

Confidence signalAvailable

Tool callingStructured outputReasoning controlLog probabilities

Cost per 1,000 requests

Request shape	List cost
Classification or routing (600 tokens in, 10 out)	$0.0158
Chat reply (2,000 in, 400 out)	$0.17
Retrieval-augmented answer (8,000 in, 500 out)	$0.328

Before prompt caching. Try your own shape in the cost calculator.

Can Gimbix route to it first?

Yes. 16 of 31 tracked hosts return token log probabilities for DeepSeek V4 Flash 0731, which is the confidence signal Gimbix uses to let a cheaper model answer first.

Measured by Gimbix

On 400 real BANKING77 customer messages, DeepSeek V4 Flash 0731 answered 298 correctly (95% interval 70.0%–78.5%), with 1 invalid answers, at $0.0101 per 1,000 requests and a median of 1,035 ms. See all 22 models.

Who serves DeepSeek V4 Flash 0731

Host	Input / output per 1M	Log probabilities	Uptime, last 30 min
Relace	$0.021 / $0.32	no	99.9%
Reka	$0.021 / $0.528	yes	99.8%
Sail Research	$0.0215 / $0.30	yes	99.5%
Sail Research	$0.0225 / $0.42	yes	99.8%
OpenInference	$0.03 / $0.40	no	100.0%
Inceptron	$0.039 / $0.396	yes	99.9%
StreamLake	$0.044 / $0.132	yes	99.5%
Wafer	$0.059 / $0.35	yes	100.0%
DeepInfra	$0.06 / $0.18	no	99.8%
Makora	$0.09 / $0.195	yes	99.9%
DigitalOcean	$0.119 / $0.238	yes	100.0%
BaseTen	$0.13 / $0.26	no	100.0%
BaseTen	$0.13 / $0.26	no	100.0%
CoreWeave	$0.13 / $0.28	yes	100.0%
Nebius	$0.14 / $0.28	no	95.1%
Cohere	$0.14 / $0.28	yes	99.8%
Together	$0.14 / $0.28	no	99.9%
Parasail	$0.14 / $0.28	yes	99.2%
Morph	$0.142 / $0.40	yes	99.7%
Venice	$0.175 / $0.35	no	100.0%
Mancer 2	$0.20 / $0.60	yes	98.9%
Fireworks	$0.22 / $0.66	yes	0.0%
SiliconFlow	$0.22 / $0.66	no	99.4%
GMICloud	$0.286 / $0.858	yes	99.8%
NextBit	$0.352 / $1.06	no	100.0%
Alibaba	$0.352 / $1.06	no	99.8%
Novita	$0.409 / $1.23	yes	100.0%
Phala	$0.44 / $1.32	no	99.9%
Baidu	$0.44 / $1.32	no	100.0%
AtlasCloud	$0.44 / $1.32	no	99.8%
Cloudflare	$0.44 / $1.32	yes	100.0%

A host that lists log probabilities may still omit them; Gimbix checks each one before relying on it.

Cheaper models that can answer first

Models priced below DeepSeek V4 Flash 0731 that expose a confidence signal. Whether they answer your traffic as well is what a Gimbix pilot measures.

Model	Input / output per 1M	Cheaper on a chat reply
Granite 4.0 Micro	$0.017 / $0.112	54%
gpt-oss-20b	$0.018 / $0.09	58%
Mistral Nemo	$0.019 / $0.03	71%

Compare DeepSeek V4 Flash 0731

Questions

How much does DeepSeek V4 Flash 0731 cost?

DeepSeek V4 Flash 0731 lists at $0.021 per million input tokens and $0.32 per million output tokens on OpenRouter as of 2026-09-28. A typical chat reply of 2,000 input and 400 output tokens costs about $0.17 per 1,000 requests.

What is the context window of DeepSeek V4 Flash 0731?

1,310,720 tokens.

Can a cheaper model replace DeepSeek V4 Flash 0731?

Sometimes. Gimbix proves it on your own traffic before routing: the cheaper model must match DeepSeek V4 Flash 0731 on one half of your requests and make at most three answers worse on the other.