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

Qwen3 30B A3B

alibaba· 2025-04-28· 41K ctx· apache-2.0
open weights quant: unknown

spread (output)

3.7×

$0.290 – $1.08 · 12 providers

12 providers priced
ProviderInputOutputCache rdBlended
Helicone
relay router
$0.080$0.290—$0.099*
NanoGPT
relay router
$0.100$0.300$0.050$0.086
Cloudflare Workers AIfp8
relay router
$0.051$0.335—$0.077*
NovitaAIfp8
relay router
$0.090$0.450—$0.123*
Jiekou.AIfp8
relay router
$0.090$0.450—$0.123*
Deep Infra
open weights host
$0.120$0.500—$0.155*
Vercel AI Gateway
unknown
$0.120$0.500—$0.155*
OpenRouter
unknown
$0.120$0.500—$0.155*
Hugging Face
relay router
$0.120$0.500—$0.155*
Kilo Gateway
relay router
$0.130$0.520—$0.165*
Merge Gatewayoutlier
unknown
$0.108$1.08$0.022$0.141
302.AIoutlier
relay router
$0.110$1.08—$0.198*

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