Gemini 2.5 Pro

Google · Released 17 Jun 2025 · gemini-2.5-pro

Compare this model

ProprietaryExisting users onlyCodingReasoning

$1.25

Input price / 1M tokens

$10.00

Output price / 1M tokens

1.05M

Context window

#11 of 71

122

Output tokens / second

#24 of 61

21.6%

Humanity’s Last Exam (no tools)

#16 of 23

86.4%

GPQA Diamond

#19 of 27

88%

AIME 2025

#6 of 7

68%

MMMU-Pro

#7 of 12

Summary

Gemini 2.5 Pro is a proprietary model from Google, released on 17 Jun 2025.

Its 1.05M-token context window ranks #11 of 71.

Measured output speed is 122 tokens per second, #24 of 61.

Gemini 2.5 thinking model for complex reasoning in code, math and STEM. It is still served but closed to new users; Google points new projects to 3.5 Flash-Lite or 3.8 Flash.

Benchmarks · 8 reported

Benchmark	Score	Rank	Source
Humanity’s Last Exam (no tools)no tools, full set text + multimodal, thinking	21.6%	#16 of 23	Google Gemini 3 Flash model card (Gemini 2.5 Pro column, Dec 2025)Self-reported
GPQA Diamondno tools	86.4%	#19 of 27	Google Gemini 3 Flash model card (Gemini 2.5 Pro column, Dec 2025)Self-reported
AIME 2025no tools	88%	#6 of 7	Google Gemini 3 Flash model card (Gemini 2.5 Pro column, Dec 2025)Self-reported
MMMU-Pro	68%	#7 of 12	Google Gemini 3 Flash model card (Gemini 2.5 Pro column, Dec 2025)Self-reported
SWE-bench Verifiedsingle attempt	59.6%	#12 of 12	Google Gemini 3 Flash model card (Gemini 2.5 Pro column, Dec 2025)Self-reported
Terminal-Bench 2.0Terminus 2 harness	32.6%	#6 of 7	Google Gemini 3 Flash model card (Gemini 2.5 Pro column, Dec 2025)Self-reported
SimpleQA Verified	54.5%	–	Google Gemini 3 Flash model card (Gemini 2.5 Pro column, Dec 2025)Self-reported
MRCR v2 (8-needle)128k average (16.4% at 1M pointwise)	58%	#10 of 13	Google Gemini 3 Flash model card (Gemini 2.5 Pro column, Dec 2025)Self-reported

Self-reported means the lab ran the test itself. A rank appears only where several labs report the same version of a benchmark.

Similar models

Models with the most benchmarks in common with Gemini 2.5 Pro, and the closest scores.

Compare 4 side by side

Humanity’s Last Exam (no tools)

#16 of 23

Expert questions, no search or code

GPQA Diamond

#19 of 27

Graduate-level science questions

AIME 2025

#6 of 7

Competition mathematics

MMMU-Pro

#7 of 12

Hard questions that need an image

Compare Gemini 2.5 Pro side by side

Pre-filled with the two closest models. Swap any of them.

Open comparison

Sources

price, long-context tier, cache, batch, free tier: ai.google.dev/gemini-api/docs/pricing, checked 2 Oct 2026
context, max output, knowledge cutoff, access limit: ai.google.dev/gemini-api/docs/models/gemini-2.5-pro, checked 2 Oct 2026
not deprecated, limited access: ai.google.dev/gemini-api/docs/models, checked 2 Oct 2026
release date, no shutdown date: ai.google.dev/gemini-api/docs/deprecations, checked 2 Oct 2026
benchmarks: storage.googleapis.com/deepmind-media/Model-Cards/Gemini-3-Flash-Model-Card.pdf, checked 2 Oct 2026
output speed: artificialanalysis.ai/models/gemini-2-5-pro, checked 2 Oct 2026

Building on Gemini 2.5 Pro?

Get the architecture right before the bill arrives.

We size caching, routing and fallbacks for your workload on a free call, and tell you where a cheaper model is good enough.

Book a free call All AI models