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

DeepSeek-V3

deepseek· 2024-12-26· 164K ctx
open weightstool callstructured quant: unknown

spread (output)

2.8×

$0.890 – $2.50 · 14 providers

14 providers priced
ProviderInputOutputCache rdBlended
iFlow
relay router
free/—free/——free/—
Deep Infra
open weights host
$0.320$0.890—$0.372*
Eden AI
relay router
$0.320$0.890—$0.372*
SiliconFlow (China)
relay router
$0.250$1.00—$0.318*
SiliconFlow
relay router
$0.250$1.00—$0.318*
SiliconFlow (China)
relay router
$0.250$1.00—$0.318*
D.Run (China)
relay router
$0.280$1.10—$0.355*
Pioneer
relay router
$0.270$1.12$0.135$0.284
Alibaba (China)
relay router
$0.287$1.15—$0.365*
Together AI
open weights host
$1.25$1.25—$1.25*
NovitaAI
relay router
$0.400$1.30—$0.482*
Hugging Face
relay router
$0.400$1.30—$0.482*
Merge Gateway
unknown
$0.580$1.68—$0.680*
Helicone
relay router
$0.560$1.68$0.070$0.350
NanoGPToutlier
relay router
$1.10$2.50$0.040$0.553

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