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

qwen/qwen-2.5-72b-instruct

qwen· 131K ctx
open weightstool call quant: unknown

spread (output)

15×

$0.380 – $5.60 · 9 providers

9 providers priced
ProviderInputOutputCache rdBlended
Abacus
relay router
$0.110$0.380—$0.135*
OpenRouter
unknown
$0.360$0.400—$0.364*
NovitaAI
relay router
$0.380$0.400—$0.382*
Kilo Gateway
relay router
$0.360$0.400—$0.364*
NanoGPT
relay router
$0.357$0.408$0.178$0.248
SiliconFlow
relay router
$0.590$0.590—$0.590*
SiliconFlow (China)
relay router
$0.590$0.590—$0.590*
Alibaba (China)outlier
relay router
$0.574$1.72—$0.678*
Alibabaoutlier
first party lab
$1.40$5.60—$1.78*

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