Enterprise DNAEnterprise DNA
DeepSeek logo

Models · Best cheap

DeepSeek V3.2

by DeepSeek

DeepSeek-V3.2 is a large language model designed to harmonize high computational efficiency with strong reasoning and agentic tool-use performance.

Visit DeepSeek ↗

Our verdict

Not reviewed yet

We have not reviewed DeepSeek V3.2 yet

No one at Enterprise DNA has checked this listing against its sources. The description comes from public listings gathered automatically. Any fact below shows where it came from and when it was last checked.

How we review records

From the listing

About DeepSeek V3.2

DeepSeek-V3.2 is a large language model designed to harmonize high computational efficiency with strong reasoning and agentic tool-use performance.

Best for

High-volume cheap coding and chat

Use cases

  • coding assistants
  • batch analysis
  • draft generation

Enterprise DNA angle

A/B DeepSeek V3 against Sonnet inside Omni with shared eval harnesses.

Notes

DeepSeek V3.2 is listed in the EDNA Models directory from the OpenRouter catalogue.

Pros

  • Strong quality at aggressive pricing
  • Useful coding and general assistant workloads

Cons

  • Validate safety/brand tone before customer-facing use
  • Ecosystem tooling less mature than US labs

Key facts

What we have checked

Input price
$0.260 per million tokens
Checked OpenRouter models API
Output price
$0.380 per million tokens
Checked OpenRouter models API
Price on atlas-cloud
$0.130 per million cached input tokens
Checked OpenRouter models API
Price on atlas-cloud
$0.260 per million input tokens
Checked OpenRouter models API
Price on atlas-cloud
$0.380 per million output tokens
Checked OpenRouter models API
Price on baidu
$0.028 per million cached input tokens
Checked OpenRouter models API
Price on baidu
$0.280 per million input tokens
Checked OpenRouter models API
Price on baidu
$0.420 per million output tokens
Checked OpenRouter models API
Price on deepinfra
$0.130 per million cached input tokens
Checked OpenRouter models API
Gpqa diamond
83.4% (for 1 Dec 2025)
Checked Epoch AI (CC BY 4.0)
Terminal bench
39.6% (for 1 Dec 2025)
Checked Epoch AI (CC BY 4.0)
Arc agi 2
4% (for 1 Dec 2025)
Checked Epoch AI (CC BY 4.0)

Each value shows when we last confirmed it and where it came from. Values we have not confirmed are left out, never guessed.

API access

How to call it

Call via OpenRouter with model id `deepseek/deepseek-v3.2`.

  • openrouter · deepseek/deepseek-v3.2 · Docs

OpenRouter id: deepseek/deepseek-v3.2

Price

What DeepSeek V3.2 costs, over time

Input price

$0.269 to $0.260

US dollars per million input tokens. 3 Sept 2026 to 10 Nov 2026, 44 days.

Source: OpenRouter models API. Checked .

Output price

$0.400 to $0.380

US dollars per million output tokens. 3 Sept 2026 to 10 Nov 2026, 44 days.

Source: OpenRouter models API. Checked .

Prices by host

HostPriceOver timeChecked
atlas-cloudCached input$0.130 per million cached input tokensSince 24 Oct 2026No change recordedOpenRouter models API
atlas-cloudInput$0.260 per million input tokensSince 24 Oct 2026No change recordedOpenRouter models API
atlas-cloudOutput$0.380 per million output tokensSince 24 Oct 2026No change recordedOpenRouter models API
baiduCached input$0.028 per million cached input tokensSince 24 Oct 2026No change recordedOpenRouter models API
baiduInput$0.280 per million input tokensSince 24 Oct 2026No change recordedOpenRouter models API
baiduOutput$0.420 per million output tokensSince 24 Oct 2026No change recordedOpenRouter models API
deepinfraCached input$0.130 per million cached input tokensSince 24 Oct 2026No change recordedOpenRouter models API
deepinfraInput$0.260 per million input tokensSince 24 Oct 2026No change recordedOpenRouter models API
deepinfraOutput$0.380 per million output tokensSince 24 Oct 2026No change recordedOpenRouter models API
digitaloceanCached input$0.090 per million cached input tokensSince 24 Oct 2026No change recordedOpenRouter models API
digitaloceanInput$0.300 per million input tokensSince 24 Oct 2026No change recordedOpenRouter models API
digitaloceanOutput$0.960 per million output tokensSince 24 Oct 2026No change recordedOpenRouter models API
gmicloudCached input$0.022 per million cached input tokensSince 24 Oct 2026No change recordedOpenRouter models API
gmicloudInput$0.209 per million input tokensSince 24 Oct 2026No change recordedOpenRouter models API
gmicloudOutput$0.310 per million output tokensSince 24 Oct 2026No change recordedOpenRouter models API
google-vertexInput$0.560 per million input tokensSince 24 Oct 2026No change recordedOpenRouter models API
google-vertexOutput$1.68 per million output tokensSince 24 Oct 2026No change recordedOpenRouter models API
maraInput$3.00 per million input tokensSince 24 Oct 2026No change recordedOpenRouter models API
maraOutput$4.50 per million output tokensSince 24 Oct 2026No change recordedOpenRouter models API
phalaCached input$0.500 per million cached input tokensSince 24 Oct 2026No change recordedOpenRouter models API
phalaInput$1.00 per million input tokensSince 24 Oct 2026No change recordedOpenRouter models API
phalaOutput$1.00 per million output tokensSince 24 Oct 2026No change recordedOpenRouter models API
sambanovaInput$3.00 per million input tokensSince 24 Oct 2026No change recordedOpenRouter models API
sambanovaOutput$4.50 per million output tokensSince 24 Oct 2026No change recordedOpenRouter models API
siliconflowCached input$0.140 per million cached input tokensSince 24 Oct 2026No change recordedOpenRouter models API
siliconflowInput$0.259 per million input tokensSince 24 Oct 2026No change recordedOpenRouter models API
siliconflowOutput$0.800 per million output tokensSince 24 Oct 2026No change recordedOpenRouter models API
veniceCached input$0.130 per million cached input tokensSince 24 Oct 2026No change recordedOpenRouter models API
veniceInput$0.268 per million input tokensSince 24 Oct 2026No change recordedOpenRouter models API
veniceOutput$0.390 per million output tokensSince 24 Oct 2026No change recordedOpenRouter models API

Usage

How much DeepSeek V3.2 is used

Tokens served a day

50.9B to 41.3B

tokens a day on OpenRouter. 19 Sept 2026 to 10 Nov 2026, 53 readings.

Source: OpenRouter model activity (748 of 748 series). Checked .

Scores

How DeepSeek V3.2 scores

Each score comes from the named source and shows when we last confirmed it. A flag means the score needs care.

ScoreValueSourceChecked
Artificial Analysis Intelligence Index21.5indexAs of 26 Oct 2026Artificial Analysis via OpenRouter
Artificial Analysis Coding Index44.2indexAs of 10 Nov 2026Artificial Analysis via OpenRouter
LMArena text rating (Elo)1,425EloAs of 10 Nov 2026LMArena (CC-BY-4.0), published 2026-11-09
LMArena web development rating (Elo)1,362EloAs of 10 Nov 2026LMArena (CC-BY-4.0), published 2026-11-09
Design Arena rating (Elo)1,171EloAs of 10 Nov 2026Design Arena via OpenRouter
Aider polyglot (best-across-versions)74.2percentAs of 1 Dec 2025Epoch AI (CC BY 4.0)
Apex agents (best-across-versions)21.3percentAs of 1 Dec 2025Epoch AI (CC BY 4.0)
Arc agi (best-across-versions)57percentAs of 1 Dec 2025Epoch AI (CC BY 4.0)
Arc agi 2 (best-across-versions)4percentAs of 1 Dec 2025Epoch AI (CC BY 4.0)
Cl bench (best-across-versions)12.4percentAs of 1 Dec 2025Epoch AI (CC BY 4.0)
Cl bench life (best-across-versions)7.4percentAs of 1 Dec 2025Epoch AI (CC BY 4.0)
Dtbench (best-across-versions)85.6percentAs of 1 Dec 2025Epoch AI (CC BY 4.0)
Show 7 more scores
ScoreValueSourceChecked
Frontiermath 2025 02 28 private (best-across-versions)22.1percentAs of 1 Dec 2025Epoch AI (CC BY 4.0)
Frontiermath tier 4 2025 07 01 private (best-across-versions)2.1percentAs of 1 Dec 2025Epoch AI (CC BY 4.0)
Gpqa diamond (best-across-versions)83.4percentAs of 1 Dec 2025Epoch AI (CC BY 4.0)
Lmca (best-across-versions)28.8percentAs of 1 Dec 2025Epoch AI (CC BY 4.0)
Otis mock aime 2024 2025 (best-across-versions)Saturated87.8percentAs of 1 Dec 2025Epoch AI (CC BY 4.0)
Proofbench (best-across-versions)8percentAs of 1 Dec 2025Epoch AI (CC BY 4.0)
Terminal bench (best-across-versions)39.6percentAs of 1 Dec 2025Epoch AI (CC BY 4.0)
Saturated
Top models score close to the maximum, so this test no longer separates them.

Sources

Where this comes from

A source shows what was published when we checked it. It does not certify security, performance or a working installation. Read our method.

Changelog

What changed

  1. Price changed

    output per mtok: 0.42 to 0.38

    Source: OpenRouter models API

  2. Price changed

    input per mtok: 0.28 to 0.26

    Source: OpenRouter models API

  3. Price changed

    output per mtok: 0.8 to 0.42

    Source: OpenRouter models API

  4. Price changed

    input per mtok: 0.259 to 0.28

    Source: OpenRouter models API

  5. Price changed

    output per mtok: 0.42 to 0.8

    Source: OpenRouter models API

  6. Price changed

    input per mtok: 0.28 to 0.259

    Source: OpenRouter models API

  7. Price changed

    output per mtok: 0.4 to 0.42

    Source: OpenRouter models API

  8. Price changed

    input per mtok: 0.269 to 0.28

    Source: OpenRouter models API

  9. Price changed

    output per mtok: 0.3217 to 0.4

    Source: OpenRouter models API

  10. Price changed

    input per mtok: 0.2145 to 0.269

    Source: OpenRouter models API

USAGE / OPENROUTER

How much DeepSeek V3.2 is used

Tokens processed through OpenRouter each day. It shows what developers actually run, not every company's private traffic.

Tokens, last 7 days
291B ▲4.8%
Requests, last 7 days
34.7M
Tool calls, last 7 days
1.1M
People's vote
1425
Average cost per request
15.6¢
Price per million tokens
$0.26 in · $0.38 out
050B100B19 Sept: 50.9B tokens, 8M requests20 Sept: 52.7B tokens, 7.3M requests21 Sept: 56.8B tokens, 7.9M requests22 Sept: 55.1B tokens, 7.6M requests23 Sept: 52.4B tokens, 7.4M requests24 Sept: 51.5B tokens, 8.6M requests25 Sept: 54.8B tokens, 8.9M requests26 Sept: 46.5B tokens, 6.7M requests27 Sept: 50.1B tokens, 6.9M requests28 Sept: 53.8B tokens, 7.1M requests29 Sept: 50.7B tokens, 7.7M requests30 Sept: 53.5B tokens, 7.6M requests1 Oct: 58.3B tokens, 8.7M requests2 Oct: 56.3B tokens, 9.4M requests3 Oct: 50.7B tokens, 7.1M requests4 Oct: 49B tokens, 7.2M requests5 Oct: 49.6B tokens, 7.1M requests6 Oct: 47.5B tokens, 7.7M requests7 Oct: 47.6B tokens, 7.6M requests8 Oct: 45.6B tokens, 6.6M requests9 Oct: 46.4B tokens, 6.5M requests10 Oct: 44.3B tokens, 8.3M requests11 Oct: 53.5B tokens, 10.3M requests12 Oct: 42.5B tokens, 5.6M requests13 Oct: 41.5B tokens, 6.3M requests14 Oct: 41.1B tokens, 5.8M requests15 Oct: 40.1B tokens, 6.1M requests16 Oct: 41.7B tokens, 6.1M requests17 Oct: 44.1B tokens, 6.1M requests18 Oct: 42.9B tokens, 5.7M requests19 Oct: 39.4B tokens, 5.7M requests20 Oct: 38.1B tokens, 5.5M requests21 Oct: 36.4B tokens, 5.1M requests22 Oct: 38.7B tokens, 5.7M requests23 Oct: 39.1B tokens, 5.5M requests24 Oct: 41.1B tokens, 6M requests25 Oct: 45.2B tokens, 5.2M requests26 Oct: 40.8B tokens, 5.5M requests27 Oct: 40.7B tokens, 5.5M requests28 Oct: 40.7B tokens, 5.2M requests29 Oct: 40.5B tokens, 5.2M requests30 Oct: 38.8B tokens, 5.2M requests31 Oct: 38.7B tokens, 5M requests1 Nov: 40.9B tokens, 4.8M requests2 Nov: 39.7B tokens, 4.8M requests3 Nov: 38.2B tokens, 4.8M requests4 Nov: 39.8B tokens, 4.8M requests5 Nov: 38.4B tokens, 4.7M requests6 Nov: 44.9B tokens, 5M requests7 Nov: 40.4B tokens, 5M requests8 Nov: 44.8B tokens, 4.9M requests9 Nov: 41.1B tokens, 5M requests10 Nov: 41.3B tokens, 5.2M requests19 Sept3 Oct17 Oct31 Oct10 NovSee where it ranks on the leaderboard →

MODEL FIELD GUIDE / SOURCE EVIDENCE

What you can do with DeepSeek V3.2.

Listed specifications; no EDNA task evaluation has been run. Sources checked 2026-11-06. Hosting routes and app features can differ.

DeveloperDeepSeek
Exact model identifierdeepseek/deepseek-v3.2
Inputtext
Outputtext
Tool calling
Listed support
Structured output
Listed support
Reasoning controls
Listed support
Availability
available
Price basis
token
Licence verification
MIT, open weights
Context length
164K tokens
Knowledge cutoff
Not established
Developer release date
Not established
First seen in the catalogue
1 Dec 2025
Listed retirement date
Not established
Inspect the sources and their scope
catalogue source ↗

deepseek/deepseek-v3.2: listed route capabilities, availability and prices; provider endpoints can differ.

Checked 2026-11-06
openrouter.ai ↗

Source for reasoning controls, first seen, tool calling, context length, input, output, structured output.

Checked 2026-11-06
Licence source ↗

Licence declared in the deepseek-ai/DeepSeek-V3.2 model repository on Hugging Face, which holds the weights files. Read the licence terms before use.

Checked 2026-11-07
How to evaluate a model for your work →

FROM SPECIFICATIONS TO YOUR WORK

Is this the right model for your job?

Start with a realistic sample. These workflows match the model’s listed capabilities, not a verified result on your task.

Write a clear business documentWorkflow & starter prompt ↗Analyse a spreadsheetWorkflow & starter prompt ↗Research a companyWorkflow & starter prompt ↗Build an appWorkflow & starter prompt ↗Run a reliable agentWorkflow & starter prompt ↗

What would it cost?

Pick a common job and how many times a month you run it. Token sizes are our assumption for a typical run, shown under the result.

Working out the cost...

Token prices only. Per-request, image, tool, storage and subscription fees are not included. One token is about three quarters of an English word. Real counts vary by model and language. How tokens are counted ↗

Check the source

Price date: 10 Nov 2026, from OpenRouter models API. Prices are USD API token rates. Your app subscription, uploads, browsing and tools are separate.

Model source & availability ↗

Understand the evidence

Listed capabilities and third-party benchmark scores do not establish Enterprise DNA task-test results. Validate accuracy and total cost on your own sample.

Coding Index: 44.2 (2026-11-09) ↗Design Arena: 1171 (2026-11-09) ↗
Compare this model with alternatives →
Free 27-page guide

Get the free Developer’s Field Guide

A 27-page field guide to the AI coding workflow with Claude. Claude Code, MCP servers, the prompt patterns that work, and what to delegate. Free.

Enter your work email. We send it straight over, plus a few short notes worth knowing. Unsubscribe any time.

No spam. Unsubscribe any time.

← Back to ModelsSubmit your own entry →