opendunes

DeepSeek V4 Flash 0731

by DeepSeekReleased 31 July 2026

DeepSeek's generally available V4 Flash, re-post-trained for coding, reasoning and agent work on the same 284B mixture-of-experts with 13B active, with tools, structured output and switchable reasoning

TakesText
GivesText
Open in chatCreate API keyCompare

What it costs

You pay for what each request uses. No subscription, no minimum.

≈ 0,070 DA per message3 000 DA buys about 43 000 messages

One message, the ordinary use

1 000 tokens sent× 19,64 DA per 1M0,020 DA
500 tokens back× 99,85 DA per 1M0,050 DA
Total≈ 0,070 DA

Every rate

Meter↑ you send↓ comes back
Text tokens19,64 DA99,85 DA
Cached tokens read11,65 DA

per 1M tokens

Answer a quick question≈ 0,028 DA
Reply to a customer email≈ 0,052 DA
Explain a stack trace and suggest a fix≈ 0,089 DA
Translate a full page of text≈ 0,094 DA
Review a 300-line pull request≈ 0,17 DA
When you pay
After the answer, from your balance. Nothing is held.
Why it varies
A longer question or a longer answer costs more.
After each one
The exact charge shows on the request.
DepositDeposits from 50 DA, with Edahabia or CIB.

Where your request runs

25 companies run this model. OpenDunes sends each request to one of them, picked by price and speed. If it fails, the next one takes it. DA per 1M tokens: ↑ what you send · ↓ what comes back.

Baidu89 % off…
↑ Sent, per 1M
Was 146,4515,67 DA
↓ Back, per 1M
Was 439,3547,01 DA
Cache read
Was 4,670,50 DA
Latency
…
Throughput
…
WaferUnited States…
↑ Sent, per 1M
19,64 DA
↓ Back, per 1M
99,85 DA
Cache read
11,65 DA
Latency
…
Throughput
…
DeepInfraUnited States…
↑ Sent, per 1M
19,97 DA
↓ Back, per 1M
59,91 DA
Cache read
4,99 DA
Latency
…
Throughput
…
StreamLake80 % off…
↑ Sent, per 1M
Was 146,4529,29 DA
↓ Back, per 1M
Was 439,3587,87 DA
Cache read
Was 4,650,93 DA
Latency
…
Throughput
…
PrimeIntellect…
↑ Sent, per 1M
29,96 DA
↓ Back, per 1M
76,55 DA
Cache read
9,65 DA
Latency
…
Throughput
…
Wafer · Fast…
↑ Sent, per 1M
33,28 DA
↓ Back, per 1M
76,55 DA
Cache read
31,62 DA
Latency
…
Throughput
…
Reka…
↑ Sent, per 1M
33,28 DA
↓ Back, per 1M
93,19 DA
Cache read
13,31 DA
Latency
…
Throughput
…
DigitalOcean…
↑ Sent, per 1M
39,61 DA
↓ Back, per 1M
79,21 DA
Cache read
7,92 DA
Latency
…
Throughput
…
BaseTenUnited States…
↑ Sent, per 1M
43,27 DA
↓ Back, per 1M
86,54 DA
Cache read
9,32 DA
Latency
…
Throughput
…
BaseTen…
↑ Sent, per 1M
43,27 DA
↓ Back, per 1M
86,54 DA
Cache read
9,32 DA
Latency
…
Throughput
…
VeniceUnited States…
↑ Sent, per 1M
43,27 DA
↓ Back, per 1M
86,54 DA
Cache read
9,32 DA
Latency
…
Throughput
…
CoreWeaveUnited States…
↑ Sent, per 1M
43,27 DA
↓ Back, per 1M
89,87 DA
Cache read
16,64 DA
Latency
…
Throughput
…
InceptronEurope…
↑ Sent, per 1M
43,27 DA
↓ Back, per 1M
164,75 DA
Cache read
8,99 DA
Latency
…
Throughput
…
Nebius…
↑ Sent, per 1M
46,60 DA
↓ Back, per 1M
93,19 DA
Cache read
3,33 DA
Latency
…
Throughput
…
Cohere…
↑ Sent, per 1M
46,60 DA
↓ Back, per 1M
93,19 DA
Cache read
23,30 DA
Latency
…
Throughput
…
Parasail…
↑ Sent, per 1M
46,60 DA
↓ Back, per 1M
93,19 DA
Cache read
16,64 DA
Latency
…
Throughput
…
MorphUnited States…
↑ Sent, per 1M
47,26 DA
↓ Back, per 1M
133,13 DA
Cache read
11,98 DA
Latency
…
Throughput
…
Sail ResearchUnited States…
↑ Sent, per 1M
49,93 DA
↓ Back, per 1M
99,85 DA
Cache read
13,98 DA
Latency
…
Throughput
…
Sail Research…
↑ Sent, per 1M
49,93 DA
↓ Back, per 1M
99,85 DA
Cache read
13,98 DA
Latency
…
Throughput
…
Mancer 2…
↑ Sent, per 1M
58,25 DA
↓ Back, per 1M
166,42 DA
Cache read
58,25 DA
Latency
…
Throughput
…
AlibabaAsia…
↑ Sent, per 1M
58,58 DA
↓ Back, per 1M
175,74 DA
Cache read
5,86 DA
Latency
…
Throughput
…
AlibabaUnited States…
↑ Sent, per 1M
62,24 DA
↓ Back, per 1M
186,72 DA
Cache read
6,22 DA
Latency
…
Throughput
…
SiliconFlowUnited States…
↑ Sent, per 1M
73,22 DA
↓ Back, per 1M
219,67 DA
Cache read
9,32 DA
Latency
…
Throughput
…
GMICloud35 % offUnited States…
↑ Sent, per 1M
Was 146,4595,19 DA
↓ Back, per 1M
Was 439,34285,57 DA
Cache read
Was 4,663,03 DA
Latency
…
Throughput
…
Phala30 % off…
↑ Sent, per 1M
Was 146,44102,51 DA
↓ Back, per 1M
Was 439,34307,54 DA
Cache read
Was 9,316,52 DA
Latency
…
Throughput
…
Novita7 % off…
↑ Sent, per 1M
Was 146,45136,20 DA
↓ Back, per 1M
Was 439,34408,59 DA
Cache read
Was 9,328,67 DA
Latency
…
Throughput
…
Chutes…
↑ Sent, per 1M
146,45 DA
↓ Back, per 1M
439,34 DA
Cache read
14,64 DA
Latency
…
Throughput
…
AtlasCloud…
↑ Sent, per 1M
146,45 DA
↓ Back, per 1M
439,34 DA
Cache read
9,32 DA
Latency
…
Throughput
…
Cloudflare…
↑ Sent, per 1M
146,45 DA
↓ Back, per 1M
439,34 DA
Cache read
4,66 DA
Latency
…
Throughput
…
Provider↑ Sent↓ BackCache readLatencyThroughputUptime
Baidu89 % offWas 146,4515,67 DAWas 439,3547,01 DAWas 4,670,50 DA………
WaferUnited States19,64 DA99,85 DA11,65 DA………
DeepInfraUnited States19,97 DA59,91 DA4,99 DA………
StreamLake80 % offWas 146,4529,29 DAWas 439,3587,87 DAWas 4,650,93 DA………
PrimeIntellect29,96 DA76,55 DA9,65 DA………
Wafer · Fast33,28 DA76,55 DA31,62 DA………
Reka33,28 DA93,19 DA13,31 DA………
DigitalOcean39,61 DA79,21 DA7,92 DA………
BaseTenUnited States43,27 DA86,54 DA9,32 DA………
BaseTen43,27 DA86,54 DA9,32 DA………
VeniceUnited States43,27 DA86,54 DA9,32 DA………
CoreWeaveUnited States43,27 DA89,87 DA16,64 DA………
InceptronEurope43,27 DA164,75 DA8,99 DA………
Nebius46,60 DA93,19 DA3,33 DA………
Cohere46,60 DA93,19 DA23,30 DA………
Parasail46,60 DA93,19 DA16,64 DA………
MorphUnited States47,26 DA133,13 DA11,98 DA………
Sail ResearchUnited States49,93 DA99,85 DA13,98 DA………
Sail Research49,93 DA99,85 DA13,98 DA………
Mancer 258,25 DA166,42 DA58,25 DA………
AlibabaAsia58,58 DA175,74 DA5,86 DA………
AlibabaUnited States62,24 DA186,72 DA6,22 DA………
SiliconFlowUnited States73,22 DA219,67 DA9,32 DA………
GMICloud35 % offUnited StatesWas 146,4595,19 DAWas 439,34285,57 DAWas 4,663,03 DA………
Phala30 % offWas 146,44102,51 DAWas 439,34307,54 DAWas 9,316,52 DA………
Novita7 % offWas 146,45136,20 DAWas 439,34408,59 DAWas 9,328,67 DA………
Chutes146,45 DA439,34 DA14,64 DA………
AtlasCloud146,45 DA439,34 DA9,32 DA………
Cloudflare146,45 DA439,34 DA4,66 DA………

You pay the price of the provider that served your request. Prices read 15:19 · speed over the last 30 minutes · uptime over 3 days.

See it in action

Real requests, real replies. Try any of them yourself — one click, pre-filled.

Structured output
You send

Turn this customer email into JSON with intent, urgency and order_id.

customer-email.txt
It replied

{ "intent": "refund", "urgency": "high", "order_id": "OD-58421" }

Try in Chat Lab
Writing
You send

Two lines to welcome a shop in Oran.

It replied

Welcome to our new shop in Oran — fresh bread every morning, and coffee the way the city likes it. Come in, the first cup is on us.

Try in Chat Lab

Try your own request — same model, pre-selected.

Open Chat Lab

Call it in code

Swap the base URL and the key. Your OpenAI code runs unchanged.

POST /v1/chat/completions
Base URL
https://opendunes.com/v1
Model ID
deepseek/deepseek-v4-flash-0731
import os
from openai import OpenAI

client = OpenAI(
    base_url="https://opendunes.com/v1",
    api_key=os.environ["OPENDUNES_API_KEY"],
)

resp = client.chat.completions.create(
    model="deepseek/deepseek-v4-flash-0731",
    messages=[{"role": "user", "content": "Two lines to welcome a shop in Oran."}],
)

print(resp.choices[0].message.content)
pip install openaiFull reference for this route
Sign in to run it

Runs this request once on your balance.

Parameters you can send

Standard OpenAI fields, grouped by what they control. Anything not listed is accepted and ignored.

Shaping the reply

How long, how varied, where to stop.

max_tokenstemperaturetop_ptop_kmin_ptop_astopseedfrequency_penaltypresence_penaltyrepetition_penaltylogit_biaslogprobstop_logprobs

Reasoning

Let it think before answering — and see the thinking.

reasoningreasoning_effortinclude_reasoning

Structured output

Get JSON you can parse instead of prose.

response_formatstructured_outputs

Tools

Let the model call your functions.

toolstool_choiceparallel_tool_calls

Questions about DeepSeek V4 Flash 0731

  • How much does DeepSeek V4 Flash 0731 cost in dinars?

    About 0,070 DA per message. ↑ 19,64 DA · ↓ 99,85 DA per 1M tokens.

  • Is DeepSeek V4 Flash 0731 free? Can I try it?

    No. You pay per use from your balance, with no subscription. Try it in Chat at the same price.

  • What is DeepSeek V4 Flash 0731 best for?

    DeepSeek's generally available V4 Flash, re-post-trained for coding, reasoning and agent work on the same 284B mixture-of-experts with 13B active, with tools, structured output and switchable reasoning.

  • DeepSeek V4 Flash 0731 or DeepSeek V4 Flash Vision Exp?

    DeepSeek V4 Flash Vision Exp costs 0,30 DA more for the same message.

  • How do I pay for DeepSeek V4 Flash 0731 without an international card?

    Deposit dinars with your Edahabia or CIB card. Every model, DeepSeek V4 Flash 0731 included, is paid from that balance. No foreign card, no conversion.

  • Does DeepSeek V4 Flash 0731 work in Arabic, French or Darja?

    DeepSeek names no languages for DeepSeek V4 Flash 0731. Arabic, French and Darja: not stated.

  • Can it read PDFs, images or a whole book?

    It reads text. Up to 1M tokens in one request, about 1 000 pages. It writes up to 708 000 words at once.

  • Is DeepSeek V4 Flash 0731 open source?

    Yes, its weights are public. Open weights on Hugging Face

  • Who makes DeepSeek V4 Flash 0731, and when was it released?

    DeepSeek. Released 31 July 2026.

Other DeepSeek models, priced in dinars

See all DeepSeek models
DeepSeek V4.1 Flash 1017DeepSeek · Oct 17, 2026

#10 for 3D scenes · This is the 17 October 2026 snapshot of V4.1 Flash, the image-and-text model DeepSeek serves under its current Flash name. It keeps about a thousand pages in mind, replies up to about 393 pages, thinks first, uses your connected tools and fills in the exact form you give it. Pick it for agents and coding work that include screenshots, and when you want the newest Flash. The earlier DeepSeek V4.1 Flash is the snapshot before this one; V4 Pro 0813 is the larger model for hard reasoning.

Thinks firstUses your tools1 000 pages at once
≈0,15 DAper message
View model
DeepSeek V4.1 FlashDeepSeek · Sep 10, 2026

DeepSeek V4.1 Flash is the first DeepSeek model built on a new design, and it reads images as well as text, keeping about a thousand pages in mind. It thinks before answering, uses your connected tools and fills in the exact form you give it. Use it as the fast, everyday model for coding agents that also need to look at screenshots or diagrams.

Thinks firstUses your tools1 000 pages at once
≈0,11 DAper message
View model
DeepSeek V4 Flash 0423DeepSeek · Apr 24, 2026

DeepSeek V4 Flash is the first V4 release of the Flash size, built for fast answers over very long text. It reads about a thousand pages at once, thinks before answering when asked, uses your connected tools and fills in the exact form or table you give it. Pick it to run an agent or summarise a whole codebase quickly. DeepSeek V4 Flash 0731 is the retrained revision for coding and agents; V4 Pro is the larger model.

Thinks firstUses your tools1 000 pages at once
≈0,061 DAper message
View model
DeepSeek V4 Pro 0813DeepSeek · Aug 12, 2026

DeepSeek V4 Pro 0813 is the general-availability release of V4 Pro, the large model for hard reasoning, coding and long agent runs. It reads about a thousand pages of text, replies up to about 384 pages, uses your connected tools and fills in the exact form you give it, with three thinking levels (low, high and max) so you choose how much it reasons.

#28 of 119 for codingThinks firstUses your tools
≈0,81 DAper message
View model