Gemini 2.5 Pro Preview 06-05

Name: Gemini 2.5 Pro Preview 06-05
Brand: google
Price: 1.25 USD
Availability: OnlineOnly

Gemini:

2.5 Pro Preview 06-05

Google's Gemini 2.5 Pro, a reasoning model that thinks through a request before answering. Aimed at coding, mathematics and scientific work, and it held first place on the LMArena leaderboard at release.

Ranks#1Character Creation#1Character Development#1Speculation & Futurism#3Statistics#4API Documentation#4Code Review

Jailbreak1 of 9levels resisted

Website OpenRouter Docs Blog

Provider: Google AI
Released: 2025-06-05
Parameters: Not disclosed
Free API (OpenRouter): No
Per 1M tokens: $1.25 in · $10 out
Estimated tab: ~$1.08; Visible-output estimate. Thinking may cost extra.

Free API availability

No free endpoint for Gemini 2.5 Pro Preview 06-05 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

LMArena

#1 (as of June 2025)

source ↗

SWE-bench Verified

67.2%

source ↗

Get API accessProvider and language code samples

2 hostsInOutContextUptime

Google Vertex AI$1.25 in$10.00 out1M99.3% up Google AI Studio$1.25 in$10.00 out1M96% up

Per million tokens. Prices and uptime via OpenRouter, checked 13 Sep 2026.

Provider

fromimport openai  OpenAI

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

response = client.chat.completions.create(
"google/gemini-2.5-pro-preview"    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 Google Vertex AI

Deep analysisPsychometrics, taste index, writing DNA

SubjectiveBench

Taste Index

Across 44 scored outputs

How this is measured →

270.27xFloor

0100headroom →

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

Craft37

Originality20

Plays it safe

share of outputs that are the default answer

55%

Writing DNA

Stylometric Fingerprint

Based on 20 text responses

Tick = global average

Vocabulary Diversity51%

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

Sentence Length15.2 words

Average words per sentence.

Hedging0.31

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

Bold Formatting4.8

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

List Usage4.1

Bullet and numbered list items per 1,000 characters.

Section Structure0.98

Markdown headings per 1,000 characters.

Emoji Usage0.00

Emoji per 1,000 characters.

Transitions0.05

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

Opening Habits

Starts with greeting (55%)

Consistency

84%

Across 20 responses

Favorites

Movie

200% Wolf

20246.7

Album

THE

tricot

Book

The Hitchhiker's Guide to the Galaxy

Douglas Adams

City

Kyoto

Japan

Game

Portal

Action, Puzzle4.5

Model Responses

40 outputs · generated before accounting joined the chat

Reviewing agent-written code?See a Brief PR report

Keep exploring

COMPARE