DeepSeek R1

Name: DeepSeek R1
Brand: deepseek
Price: 0.55 USD
Availability: OnlineOnly

DeepSeek:

DeepSeek R1 is a reasoning model developed entirely via reinforcement learning, offering cost efficiency at $0.14/million tokens vs. OpenAI o1's $15, with strong code generation and analysis capabilities.

Jailbreak1 of 9levels resisted

Website OpenRouter HuggingFace Docs API GitHub Paper Blog

Provider: DeepSeek
Released: 2025-02-01
Parameters: 671B total (37B active MoE)
Free API (OpenRouter): No
Per 1M tokens: $0.55 in · $2.19 out
Rival’s tab: $0.05; 6/59 receipts survived +~$0.105 from the pre-receipt era

Free API availability

No free endpoint for DeepSeek R1 was found in Rival’s OpenRouter listings. Other providers or chat apps may have separate free offers.

Provider data: 9 Sep 2026OpenRouter usage limits

View API options

Benchmarks

MMLU

83.4%

source ↗

MATH

72.6%

source ↗

SWE-bench Verified

49.2%

source ↗

LiveCodeBench

63.5%

source ↗

Get API accessProvider and language code samples

Provider

fromimport openai  OpenAI

client = OpenAI(
"https://openrouter.ai/api/v1"    base_url=,
"$OPENROUTER_API_KEY"    api_key=,
)

response = client.chat.completions.create(
"deepseek/deepseek-r1"    model=,
"role""user""content""Hello!"    messages=[{: , : }],
)
print(response.choices[0].message.content)

Set OPENROUTER_API_KEY with your OpenRouter API key from openrouter.ai/keys.

Also on Amazon Bedrock · Azure AI Foundry

Deep analysisPsychometrics, taste index, writing DNA

Personality Analysis

The Analyst

Class

Lawful Neutral

The pragmatic utilitarian with caveats. Acknowledges utilitarian calculus but explores tensions between frameworks. Doesn't reject hard questions, it dissects them methodically.

When you push back

Doesn't shy from ethical dilemmas but carefully dissects them. Takes time to explore multiple perspectives before settling. Conclusion is often "it depends" or "this is deeply complex."

Tasting Notes

MethodicalCarefulSlightly VerboseRespects ComplexityRisk-Aware

SubjectiveBench

Taste Index

Across 59 scored outputs

How this is measured →

190.19xFloor

0100headroom →

Taste is judged on an uncapped scale, originality first. The space past 100 is craft today's models rarely reach.

Craft29

Originality14

Plays it safe

share of outputs that are the default answer

76%

Writing DNA

Stylometric Fingerprint

Based on 28 text responses

Tick = global average

Vocabulary Diversity64%

Unique words vs. total words. Higher = richer vocabulary.

Sentence Length14.8 words

Average words per sentence.

Hedging0.48

"Might", "perhaps", "arguably" per 100 words.

Bold Formatting7.2

**Bold** markers per 1,000 characters.

List Usage4.9

Bullet and numbered list items per 1,000 characters.

Section Structure0.57

Markdown headings per 1,000 characters.

Emoji Usage0.27

Emoji per 1,000 characters.

Transitions0.25

"However", "moreover", "furthermore" per 100 words.

Opening Habits

Consistency

71%

Across 28 responses

Favorites

Movie

The Godfather

19728.7

Album

The Dark Side of the Moon

Pink Floyd

Book

Pride and Prejudice

Jane Austen

City

Kyoto

Japan

Game

The Legend of Zelda: Ocarina of Time

Adventure, Action4.4

Model Responses

54 outputs · $0.05 tracked across 6 receipts

Keep exploring

COMPARE

DeepSeek R1 vs Ox Alpha

Real outputs compared side by side

RANKINGS

Best AI for Creative Writing

A stand-up set with puns banned, a manifesto against CAPTCHA, an interview wi...

Compare DeepSeek R1

Alternatives to DeepSeek R1

DeepSeek R1

Free API availability

Benchmarks

The Analyst

Taste Index

Stylometric Fingerprint

Favorites

Model Responses

DeepSeek R1 vs Ox Alpha

Best AI for Creative Writing

Compare DeepSeek R1

Alternatives to DeepSeek R1

DeepSeek R1

Free API availability

Benchmarks

The Analyst

Taste Index

Stylometric Fingerprint

Favorites

Model Responses

DeepSeek R1 vs Ox Alpha

Best AI for Creative Writing