modelbenchmark.io

Gemma 4 26B-A4B

Google · gemma-4-26b-a4b

Compare

Open Gemma instruction model for efficient chat and self-hosted deployments

Specification

most-agreed values

Context
262K
Max output
33K
Released
2026-04-02
Knowledge cutoff
—
Retires
—
Open weights
yes
Input
text, image, video
Output
text

Price

US dollars per million tokens · most-agreed

Input
$0.05
Output
$0.29
Cache read
$0.025

Batch and priority prices

US dollars per million tokens · per listing

PriceListing$/M
Batch inputbedrock_mantle · bedrock-mantle$0.065
Batch outputbedrock_mantle · bedrock-mantle$0.20
Priority inputbedrock_mantle · bedrock-mantle$0.2275
Priority outputbedrock_mantle · bedrock-mantle$0.70

Batch input is 50% below the interactive input price at the same listing.

Quality

2 benchmarks · 1 source

Composite
83rd
percentile of 344 scored models
Rank
60 / 344
±0.09 sd
Evidence
2 × 1
too little evidence to place with confidence
Effort range
—
one configuration only
reasoning73rd1/4 bench
math86th1/7 bench

Every score

one row per source and configuration — nothing averaged away

BenchmarkScoreConfigurationSourceRun
OTIS Mock AIME 2024-202582.2 ±5.8defaultEpoch AI2026-08-28
GPQA diamond73.2 ±3.2defaultEpoch AI2026-08-06

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

Available from 3 hosts

HostIn $/MOut $/MCache rdCache wrContextOutputRetires
empiriolabs$0.05$0.29$0.025—262K33K—
amazon-bedrock$0.13$0.40——262K33K—
bedrock_mantle · bedrock-mantle
$0.13
$0.156
$0.40
$0.48
——256K256K—

Price history

append-only observations · a listing writes a row only when its price moves

DateHostField$/M
2026-09-02bedrock_mantle · bedrock-mantleInput$0.13
2026-09-09bedrock_mantle · bedrock-mantleInput$0.13
2026-09-02bedrock_mantle · bedrock-mantleOutput$0.40
2026-09-09bedrock_mantle · bedrock-mantleOutput$0.40

More from Google

most-hosted first