Google · proprietary

Gemini 2.5 Pro

Gemini 2.5 Pro by Google appears in 2 sources with Reasoning at 35.05. Best read for Coding, High intelligence, LLM.

CreatorGoogle

Release date2025-05-20

Knowledge cutoffNot published

Context1M tokens

Input price$1.25/M tokens

Output price$10/M tokens

Modalitytext + vision

CountryUS

Metrics

All source-backed metrics

Open in leaderboard

LLM Stats Rank46ranking · source_rank

Reasoning35.05reasoning · index_reasoning

Math31.28math · index_math

Coding16.81coding · index_code

Vision22.66multimodal · index_vision

Long context29.47long_context · index_long_context

Healthcare32.41domain · index_healthcare

GPQA83 %reasoning · gpqa_score

AIME 202583 %math · aime_2025_score

SWE-bench Verified63.2 %coding · swe_bench_verified_score

Code Arena1,000 %coding · coding_arena_score

Humanity Last Exam17.8 %reasoning · hle_score

ARC-AGI v24.9 %reasoning · arc_agi_v2_score

MMMU79.6 %multimodal · mmmu_score

SimpleQA50.8 %knowledge · simpleqa_score

Context1,048,576 tokenscontext · context

Speed241.68 c/sperformance · throughput

Latency12,428.32 msperformance · latency

Input price1.25 $/Mpricing · input_price

Output price10 $/Mpricing · output_price

Arena Rating1,458.95metric

Arena Rank15metric

Vote Count117,010metric

Evidence

Citations and source overlap

Arena AILM Arena text leaderboard (overall) mirrored in the public leaderboard dataset.Retrieved 2026-05-20 LLM StatsPublic full LLM leaderboard advanced explorer snapshot, including ranking, benchmark indexes, pricing, speed, context, license, and model metadata.Retrieved 2026-06-01

FAQ

How should I read this profile?

Treat this as a source-backed model dossier, not an EvalKit-run verification. The public values are replicated from linked sources and kept source-scoped.

Is Gemini 2.5 Pro verified by EvalKit?

No. EvalKit currently shows 0 verified rows until real run evidence exists.

Why can metrics disagree?

Different sources test different tasks, dates, prompts, and aggregation methods. EvalKit keeps those differences visible instead of merging them into a fake universal score.