modelbenchmark.io

Gemma 4 26B A4B IT

Google · gemma-4-26b-a4b-it

Compare

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

Specification

google listing

Context
262K
Max output
33K
Released
2026-04-02
Knowledge cutoff
2025-04
Retires
—
Open weights
yes
Input
text, image
Output
text

Price

US dollars per million tokens · google list price

Input
$0.07
Output
$0.34
Cache read
$0.05

Rate limits

as listed

Requests per minute
300

Available from 31 hosts

HostIn $/MOut $/MCache rdCache wrContextOutputRetires
googlevendor————262K33K—
geminifreefree——262K33K—
openrouter · freefreefree——262K33K—
kilo$0.01$0.03$0.005—262K236K—
nano-gpt$0.01$0.031$0.005—262K131K—
nano-gpt · thinking$0.01$0.031$0.005—262K131K—
orcarouter$0.06$0.33$0.0075—262K33K—
openrouter$0.0675$0.225$0.0375—262K236K—
deepinfra$0.07$0.34——262K
33K
262K
—
llmgateway$0.07$0.34——262K33K—
requesty$0.07$0.34$0.07—262K33K—
hyper$0.098$0.334$0.049—256K26K—
cloudflare$0.10$0.30——256K256K—
cloudflare-workers-ai · cf$0.10$0.30$0.05—256K16K—
wandb$0.10$0.30$0.05—262K262K—
wandb · wandb$0.10$0.30$0.05—262K262K—
cortecs$0.111$0.557——262K82K—
huggingface$0.13$0.40——262K33K—
llmgateway-providers
$0.13
$0.07
$0.40
$0.34
——262K33K—
merge-gateway$0.13$0.40——262K66K—
novita$0.13$0.40——262K131K—
novita-ai$0.13$0.40——262K131K—
aihubmix · aihubmix$0.14$0.40——262K131K—
siliconflow$0.14$0.40$0.05—262K262K—
evroc$0.144$0.575——262K33K—
vercel$0.15$0.60$0.015—262K131K—
vertex_ai$0.15$0.60$0.015————
aimlapi$0.1587$0.6348$0.1587—262K128K—
scaleway · scaleway$0.25$0.50——256K33K—
scaleway$0.2804$0.5607——256K16K—
tempr————262K33K—

Price history

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

DateHostField$/M
2026-09-02nano-gptCache read$0.06
2026-10-19nano-gptCache read$0.0375
2026-10-20nano-gptCache read$0.0425
2026-10-24nano-gptCache read$0.0375
2026-10-27nano-gptCache read$0.0425
2026-10-28nano-gptCache read$0.046
2026-10-31nano-gptCache read$0.005
2026-09-02nano-gpt · thinkingCache read$0.065
2026-10-19nano-gpt · thinkingCache read$0.0375
2026-10-20nano-gpt · thinkingCache read$0.0425
2026-10-24nano-gpt · thinkingCache read$0.0375
2026-10-27nano-gpt · thinkingCache read$0.0425
2026-10-28nano-gpt · thinkingCache read$0.046
2026-10-31nano-gpt · thinkingCache read$0.005
2026-09-02hyperCache write$0.06
2026-09-02hyperCache write$0.058
2026-09-04hyperCache write$0.053
2026-09-08hyperCache write$0.055
2026-09-09hyperCache write$0.053
2026-09-02hyperInput$0.12
2026-09-02hyperInput$0.116
2026-09-04hyperInput$0.106
2026-09-08hyperInput$0.11
2026-09-09hyperInput$0.106
2026-09-11hyperInput$0.122
2026-09-12hyperInput$0.102
2026-09-15hyperInput$0.10
2026-09-16hyperInput$0.098
2026-09-02kiloInput$0.042
2026-10-31kiloInput$0.01
2026-09-02llmgateway-providersInput$0.07
2026-09-04llmgateway-providersInput$0.13
2026-09-24llmgateway-providersInput$0.07
2026-10-20llmgateway-providersInput$0.13
2026-10-21llmgateway-providersInput$0.07
2026-10-24llmgateway-providersInput$0.13
2026-10-25llmgateway-providersInput$0.07
2026-10-28llmgateway-providersInput$0.13
2026-10-31llmgateway-providersInput$0.07
2026-11-01llmgateway-providersInput$0.13
2026-09-02nano-gptInput$0.12
2026-10-19nano-gptInput$0.0675
2026-10-20nano-gptInput$0.0765
2026-10-24nano-gptInput$0.0675
2026-10-27nano-gptInput$0.0765
2026-10-28nano-gptInput$0.07
2026-10-31nano-gptInput$0.01
2026-09-02nano-gpt · thinkingInput$0.13
2026-10-19nano-gpt · thinkingInput$0.0675
2026-10-20nano-gpt · thinkingInput$0.0765
2026-10-24nano-gpt · thinkingInput$0.0675
2026-10-27nano-gpt · thinkingInput$0.0765
2026-10-28nano-gpt · thinkingInput$0.07
2026-10-31nano-gpt · thinkingInput$0.01
2026-09-02openrouterInput$0.07
2026-09-06openrouterInput$0.07
2026-09-11openrouterInput$0.042
2026-09-13openrouterInput$0.09
2026-09-17openrouterInput$0.09
2026-09-26openrouterInput$0.0675
2026-09-29openrouterInput$0.0765
2026-10-01openrouterInput$0.0765
2026-10-03openrouterInput$0.0675
2026-10-06openrouterInput$0.0765
2026-10-07openrouterInput$0.0765
2026-10-10openrouterInput$0.0675
2026-10-13openrouterInput$0.0765
2026-10-17openrouterInput$0.0675
2026-10-20openrouterInput$0.0765
2026-10-24openrouterInput$0.0675
2026-10-27openrouterInput$0.0765
2026-10-31openrouterInput$0.0675
2026-09-02openrouter · freeInputfree
2026-09-06openrouter · freeInputfree
2026-10-10openrouter · freeInputfree
2026-09-02scalewayInput$0.25
2026-10-22scalewayInput$0.2804
2026-09-02siliconflowInput$0.12
2026-10-22siliconflowInput$0.14
2026-09-02hyperOutput$0.42
2026-09-02hyperOutput$0.38
2026-09-04hyperOutput$0.368
2026-09-08hyperOutput$0.408
2026-09-09hyperOutput$0.368
2026-09-11hyperOutput$0.42
2026-09-12hyperOutput$0.356
2026-09-15hyperOutput$0.374
2026-09-16hyperOutput$0.334
2026-09-02kiloOutput$0.22
2026-10-31kiloOutput$0.03
2026-09-02llmgateway-providersOutput$0.34
2026-09-04llmgateway-providersOutput$0.40
2026-09-24llmgateway-providersOutput$0.34
2026-10-20llmgateway-providersOutput$0.40
2026-10-21llmgateway-providersOutput$0.34
2026-10-24llmgateway-providersOutput$0.40
2026-10-25llmgateway-providersOutput$0.34
2026-10-28llmgateway-providersOutput$0.40
2026-10-31llmgateway-providersOutput$0.34
2026-11-01llmgateway-providersOutput$0.40
2026-09-02nano-gptOutput$0.38
2026-10-19nano-gptOutput$0.225
2026-10-20nano-gptOutput$0.255
2026-10-24nano-gptOutput$0.225
2026-10-27nano-gptOutput$0.255
2026-10-28nano-gptOutput$0.24
2026-10-31nano-gptOutput$0.03
2026-11-01nano-gptOutput$0.031
2026-09-02nano-gpt · thinkingOutput$0.40
2026-10-19nano-gpt · thinkingOutput$0.225
2026-10-20nano-gpt · thinkingOutput$0.255
2026-10-24nano-gpt · thinkingOutput$0.225
2026-10-27nano-gpt · thinkingOutput$0.255
2026-10-28nano-gpt · thinkingOutput$0.24
2026-10-31nano-gpt · thinkingOutput$0.03
2026-11-01nano-gpt · thinkingOutput$0.031
2026-09-02openrouterOutput$0.34
2026-09-06openrouterOutput$0.34
2026-09-11openrouterOutput$0.22
2026-09-13openrouterOutput$0.30
2026-09-17openrouterOutput$0.30
2026-09-26openrouterOutput$0.225
2026-09-29openrouterOutput$0.255
2026-10-01openrouterOutput$0.255
2026-10-03openrouterOutput$0.225
2026-10-06openrouterOutput$0.255
2026-10-07openrouterOutput$0.255
2026-10-10openrouterOutput$0.225
2026-10-13openrouterOutput$0.255
2026-10-17openrouterOutput$0.225
2026-10-20openrouterOutput$0.255
2026-10-24openrouterOutput$0.225
2026-10-27openrouterOutput$0.255
2026-10-31openrouterOutput$0.225
2026-09-02openrouter · freeOutputfree
2026-09-06openrouter · freeOutputfree
2026-10-10openrouter · freeOutputfree
2026-09-02scalewayOutput$0.50
2026-10-22scalewayOutput$0.5607
2026-09-11hyperCache read$0.061
2026-09-12hyperCache read$0.051
2026-09-15hyperCache read$0.05
2026-09-16hyperCache read$0.049
2026-09-13openrouterCache read$0.05
2026-09-19openrouterCache read$0.05
2026-09-26openrouterCache read$0.0375
2026-09-29openrouterCache read$0.0425
2026-10-01openrouterCache read$0.0425
2026-10-03openrouterCache read$0.0375
2026-10-06openrouterCache read$0.0425
2026-10-07openrouterCache read$0.0425
2026-10-10openrouterCache read$0.0375
2026-10-13openrouterCache read$0.0425
2026-10-17openrouterCache read$0.0375
2026-10-20openrouterCache read$0.0425
2026-10-24openrouterCache read$0.0375
2026-10-27openrouterCache read$0.0425
2026-10-31openrouterCache read$0.0375
2026-09-29kiloCache read$0.021
2026-10-31kiloCache read$0.005
2026-10-13aimlapiCache read$0.2063
2026-10-23aimlapiCache read$0.1587
2026-10-13aimlapiInput$0.2063
2026-10-23aimlapiInput$0.1587
2026-10-13aimlapiOutput$0.8252
2026-10-23aimlapiOutput$0.6348

Other listings of this model

same model, different key

More from Google

most-hosted first