inferenstack — the market read on AI inference · last observation 2026-09-20. See the boards
Price spread

Gemma 4 26B A4B IT

google· 2026-04-02· 262K ctx· apache-2.0
open weightstool callreasoningstructured quant: unknown

spread (output)

2.7×

$0.220 – $0.600 · 21 providers

21 providers priced
ProviderInputOutputCache rdBlended
OpenRouter
unknown
free/—free/——free/—
NaN
relay router
free/—free/——free/—
Kilo Gateway
relay router
$0.042$0.220—$0.058*
Cloudflare Workers AI
relay router
$0.100$0.300—$0.118*
OpenRouter
unknown
$0.090$0.300$0.050$0.084
OrcaRouter
relay router
$0.060$0.330$0.0075$0.051
Charm Hyper
relay router
$0.098$0.334$0.049$0.088
Requesty
relay router
$0.070$0.340$0.070$0.095
Deep Infra
open weights host
$0.070$0.340—$0.095*
DevPass (LLM Gateway)
relay router
$0.070$0.340—$0.095*
LLM Gateway
relay router
$0.070$0.340—$0.095*
NanoGPT
relay router
$0.120$0.380$0.060$0.105
Amazon Bedrock
hyperscaler
$0.130$0.400—$0.155*
LLM Gateway
relay router
$0.130$0.400—$0.155*
Merge Gateway
unknown
$0.130$0.400—$0.155*
NovitaAI
relay router
$0.130$0.400—$0.155*
SiliconFlow
relay router
$0.120$0.400—$0.145*
Hugging Face
relay router
$0.130$0.400—$0.155*
AKI.IOoutlier
relay router
$0.100$0.500—$0.136*
Scalewayoutlier
relay router
$0.250$0.500—$0.273*
Cortecsoutlier
relay router
$0.111$0.557—$0.152*
evrocoutlier
relay router
$0.144$0.575—$0.183*
Vercel AI Gatewayoutlier
unknown
$0.150$0.600$0.015$0.105

Data: models.dev (MIT), snapshot 2026-09-20. Prices are vendor-published; verify before purchase. Methodology.