modelbenchmark.io

Gemma 3 27B

Google · gemma-3-27b

Compare

Largest open Gemma 3 instruction model for multilingual text generation and visual understanding

Specification

most-agreed values

Context
128K
Max output
8K
Released
2025-03-12
Knowledge cutoff
2024-08
Retires
Open weights
yes
Input
text, image
Output
text

Price

US dollars per million tokens · most-agreed

Input
free
Output
free

Quality

5 benchmarks · 2 sources

Composite
28th
percentile of 334 scored models
Rank
240 / 334
±0.06 sd
Evidence
5 × 2
3 or more benchmarks, from 2 or more sources
Effort range
one configuration only
coding5th1/3 bench
reasoning26th2/4 bench
math50th2/7 bench

Every score

one row per source and configuration — nothing averaged away

BenchmarkScoreConfigurationSourceRun
MATH level 574.0 ±1.1defaultEpoch AI2025-03-13
GPQA diamond47.7 ±2.7defaultEpoch AI2026-08-27
OTIS Mock AIME 2024-202522.5 ±5.3defaultEpoch AI2026-08-27
Chess Puzzles0.0 ±0.0defaultEpoch AI2026-08-27
Aider polyglot4.9default · aider wholeAider2025-03-15

The leading figure for a benchmark is the median across its configurations, so one heroic high-effort run cannot set the number.

Available from 1 host

HostIn $/MOut $/MCache rdCache wrContextOutputRetires
empiriolabsfreefree128K8K

Other listings of this model

this page is one of them

The corpus files Gemma 3 27B under several keys. This page is the gemma-3-27b listing. The full record — every host, every price and every benchmark score — is on the main Gemma 3 27B page.

More from Google

most-hosted first