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

Qwen3.5 35B-A3B

alibaba· 2026-02-23· 262K ctx· apache-2.0
open weights quant: unknown

spread (output)

4.4×

$0.459 – $2.00 · 20 providers

20 providers priced
ProviderInputOutputCache rdBlended
Merge Gateway
unknown
$0.057$0.459$0.020$0.070
EmpirioLabs AItiered
relay router
$0.057$0.459$0.057$0.094
OrcaRouter
relay router
$0.057$0.459—$0.094*
302.AI
relay router
$0.060$0.460—$0.096*
Deep Infra
open weights host
$0.140$1.00$0.050$0.161
Requesty
relay router
$0.140$1.00$0.050$0.161
CoreWeave
relay router
$0.250$1.25$0.250$0.341
OpenRouter
unknown
$0.313$1.25$0.156$0.298
Venice AI
unknown
$0.313$1.25$0.156$0.298
Kilo Gateway
relay router
$0.163$1.30—$0.266*
Mixlayer
relay router
$0.250$1.30—$0.346*
NanoGPT
relay router
$0.225$1.80$0.113$0.297
SiliconFlow
relay router
$0.240$1.80—$0.382*
Ofox
relay router
$0.290$1.83$0.290$0.430
Ofox
relay router
$0.290$1.83$0.290$0.430
SiliconFlow (China)
relay router
$0.230$1.86—$0.378*
Alibaba
first party lab
$0.250$2.00—$0.409*
NovitaAI
relay router
$0.250$2.00—$0.409*
Jalapeno Cloud
relay router
$0.250$2.00—$0.409*
Hugging Face
relay router
$0.250$2.00—$0.409*

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