by Google DeepMind

Gemini 1.5 Pro.

multimodal closed 1M ctx Transformer (MoE) Quality 81.9
Cheapest input
$1.25/M
on Google AI Studio
Cheapest output
$5.0/M
on Google AI Studio
Hosted equiv.
~$1.8/hr
@ 100 tok/s on Google AI Studio
Capability snapshot

What it's best at.

General knowledge 81.9

Scores normalised against benchmark ceilings (100 = perfect). Coloured by tier — coral 80+ frontier, lavender 65+ strong, sage 50+ solid, slate below.

Benchmarks

Published scores.

Benchmark Score Source
MMLU 81.9 official ↗
Description

About Gemini 1.5 Pro.

Gemini 1.5 Pro was Google's flagship for most of 2024 — introduced the 1M-token context window (later 2M) and the first widely-deployed MoE architecture from Google. Still in active use because the 1M context was a big deal and many pipelines were built around it. Superseded by Gemini 2.5 Pro in March 2025 for new builds.

Architecture

How it's built.

Architecture
Transformer (MoE)
Knowledge cutoff
Nov 2023
106 days from cutoff to release.
Context window

How much it can remember.

1M tokens ≈ 750,000 English words
4K 32K 128K 1M
Max output per call: 8K tokens
Capabilities

What it can do.

Vision input
Audio input
Video input
Function calling
· Tool use
JSON mode
Streaming
· Fine-tuning