AiCostCompare

Gemma 4 26B A4B

by Google · google/gemma-4-26b-a4b-it · #64 cheapest of 422 paid models

Prices updated Sep 19, 2026, 6:21 PM UTC · refreshed hourly

Gemma 4 26B A4B IT is an instruction-tuned Mixture-of-Experts (MoE) model from Google DeepMind. Despite 25.2B total parameters, only 3.8B activate per token during inference — delivering near-31B quality at...

Compare vs:Claude Sonnet 5Claude Opus 4.6Claude Haiku 4.5GPT-5.6 Sol

Pricing

Input

$0.09

per 1M tokens

Output

$0.30

per 1M tokens

Blended (3:1)

$0.143

3 input : 1 output

Cache read

$0.05

per 1M cached tokens

Cache write

per 1M tokens

Cache & batch economics

Effective prices for Gemma 4 26B A4B with prompt caching.

Cache hit rateEffective blended $/1MExample request*vs no cache
0%$0.143$0.00948
50%$0.128$0.00748−21%
90%$0.115$0.00588−38%

*Example: 100K sticky context + 2K new input + 1K output. Batch discount is an approximate 50% off list rates for this provider’s batch API. Confirm on the vendor’s pricing page.

What a request costs

WorkloadInput tokensOutput tokensCost / requestCost / 1K requests
Short chat message500300$0.000135$0.135
Document summary8,0001,000$0.00102$1.02
Codebase question (RAG)30,0002,000$0.0033$3.30
Long-context analysis150,0005,000$0.015$15.00

Performance

Intelligence Index

16.7

AA composite quality

Coding Index

39.3

Math Index

Agentic Index

Output speed

0 tok/s

median

Time to first token

0s

median

Time to first answer

0s

after reasoning tokens

Artificial Analysis benchmarks

BenchmarkScore
GPQA Diamond79.2%
Humanity's Last Exam19.3%
SciCode40.0%
IFBench72.4%
AA-LCR (Long Context Reasoning)65.7%
Terminal-Bench Hard13.6%
Terminal-Bench 2.139.0%
τ²-Bench (Telecom)43.6%
τ³-Bench Banking12.0%

Specs

Context window

262K

tokens

Max output

236K

tokens

Input modalities

image, text, video

Tokenizer

Gemma