Mistral Medium 3

Name: Mistral Medium 3
Brand: mistral
Price: 0.40 USD
Availability: OnlineOnly

Mistral:

Medium 3

Mistral's mid tier model, pitched at large-model quality for roughly an eighth of the cost. Strongest on coding and STEM reasoning, and deployable hybrid, on-prem or inside a VPC rather than API only.

Website OpenRouter Docs API GitHub Blog

Provider: Mistral AI
Released: 2025-05-07
Parameters: Not disclosed
Free API (OpenRouter): No
Per 1M tokens: $0.40 in · $2 out
Rival’s tab: $0.12; 27/59 receipts survived +~$0.079 from the pre-receipt era

Free API availability

No free endpoint for Mistral Medium 3 was found in Rival’s OpenRouter listings. Other providers or chat apps may have separate free offers.

Provider data: 13 Sep 2026OpenRouter usage limits

View API options

Benchmarks

HumanEval

92.1%

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(
"mistralai/mistral-medium-3"    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 SageMaker · Azure AI Foundry

Deep analysisPsychometrics, taste index, writing DNA

Personality Analysis

The Charming Generalist

Class

Neutral Good

The witty copywriter who actually finished their coffee. Treats every prompt like a creative brief, not a chore. Brings personality without being obnoxious.

When you push back

Approaches prompts with enthusiasm and delivers with polish. Standup routines actually land. Steve Jobs interviews feel genuine. Task trackers have localStorage persistence. The French model that actually ships quality, like a croissant that's also functional.

Tasting Notes

Actually FunnyArtisanal Pickle EnergySolid Landing PagesReliable Creative PartnerFrench Precision With Swagger

SubjectiveBench

Taste Index

Across 59 scored outputs

How this is measured →

170.17xFloor

0100headroom →

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

Craft28

Originality11

Plays it safe

share of outputs that are the default answer

83%

Writing DNA

Stylometric Fingerprint

Based on 28 text responses

Tick = global average

Vocabulary Diversity55%

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

Sentence Length18.0 words

Average words per sentence.

Hedging0.75

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

Bold Formatting7.0

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

List Usage5.2

Bullet and numbered list items per 1,000 characters.

Section Structure1.11

Markdown headings per 1,000 characters.

Emoji Usage0.23

Emoji per 1,000 characters.

Transitions0.16

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

Opening Habits

Consistency

71%

Across 28 responses

Favorites

Movie

The Shawshank Redemption

19948.7

Album

The Dark Side of the Moon

Pink Floyd

Book

Братья Карамазовы

Fiódor Dostoievski

City

Kyoto

Japan

Game

The Legend of Zelda: Breath of the Wild

Adventure, Action4.5

Model Responses

54 outputs · $0.12 tracked across 27 receipts

Reviewing agent-written code?See a Brief PR report

Keep exploring

COMPARE

Mistral Medium 3 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 Mistral Medium 3

Alternatives to Mistral Medium 3

Mistral Medium 3

Free API availability

Benchmarks

The Charming Generalist

Taste Index

Stylometric Fingerprint

Favorites

Model Responses

Mistral Medium 3 vs Ox Alpha

Best AI for Creative Writing

Compare Mistral Medium 3

Alternatives to Mistral Medium 3

Mistral Medium 3

Free API availability

Benchmarks

The Charming Generalist

Taste Index

Stylometric Fingerprint

Favorites

Model Responses

Mistral Medium 3 vs Ox Alpha

Best AI for Creative Writing