AiCostCompare

Gemma 4 31B

by Google · google/gemma-4-31b-it · #72 cheapest of 422 paid models

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

Gemma 4 31B Instruct is Google DeepMind's 30.7B dense multimodal model supporting text and image input with text output. Features a 256K token context window, configurable thinking/reasoning mode, native function...

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

Pricing

Input

$0.09

per 1M tokens

Output

$0.34

per 1M tokens

Blended (3:1)

$0.152

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 31B with prompt caching.

Cache hit rateEffective blended $/1MExample request*vs no cache
0%$0.152$0.00952
50%$0.137$0.00752−21%
90%$0.126$0.00592−38%

*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$0.000147$0.147
Document summary8,0001,000$0.00106$1.06
Codebase question (RAG)30,0002,000$0.00338$3.38
Long-context analysis150,0005,000$0.0152$15.20

Performance

Intelligence Index

15.4

AA composite quality

Coding Index

43.4

Math Index

Agentic Index

6.7

Output speed

35 tok/s

median

Time to first token

0.87s

median

Time to first answer

50.26s

after reasoning tokens

Artificial Analysis benchmarks

BenchmarkScore
GPQA Diamond85.7%
Humanity's Last Exam23.6%
SciCode45.5%
IFBench75.6%
AA-LCR (Long Context Reasoning)69.7%
Terminal-Bench Hard36.4%
Terminal-Bench 2.143.4%
τ²-Bench (Telecom)59.9%
τ³-Bench Banking14.8%

Specs

Context window

262K

tokens

Max output

16K

tokens

Input modalities

image, text, video

Tokenizer

Gemma