Gemini 1.0 Pro

Google

Verified against Artificial Analysis · Jul 13, 2026

Capability

percentile index

Price /M

$0 / $0

in / out

Context

—

Avg score

23.2

6 benchmarks

AA Intelligence

3.1

index

Weights

Closed

Released

Dec 6, 2023

About

Gemini 1.0 Pro is a proprietary model in the Gemini family from Google. Benchmarked on 6 evals, averaging 23.2.

Benchmark scores · 6

avg 23.2 — higher is better

MMLU-Proartificial-analysis · 2026-07

43.1

MATHartificial-analysis · 2026-07

40.3

GPQA Diamondartificial-analysis · 2026-07

27.7

SciCodeartificial-analysis · 2026-07

11.7

LiveCodeBenchartificial-analysis · 2026-07

11.6

Humanity's Last Examartificial-analysis · 2026-07

4.6

The research behind this model

all papers mentioning it →

Evaluating Frontier Models for Dangerous Capabilities10 cites · Mar 20, 2024

Mentions matched by name in title/abstract — from the arXiv + HF daily corpus.

Details

PublisherGoogle

FamilyGemini

Parameters—

License—

WeightsClosed

ReleasedDec 6, 2023

History

full ledger →

Capability over time

Append-only ledger — every observed change to this model's numbers.

API providers

Artificial Analysis$0/M