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

Nemotron 3 Ultra 550B A55B

nvidia· 2026-06-04· 1M ctx
open weightstool callreasoning quant: unknown

spread (output)

36×

$0.100 – $3.60 · 13 providers

13 providers priced
ProviderInputOutputCache rdBlended
Kilo Gateway
relay router
free/—free/——free/—
Bothub
relay router
free/—free/——free/—
OpenRouter
unknown
free/—free/——free/—
Requesty
relay router
free/—free/——free/—
Kenari
relay router
free/—free/——free/—
UnoRouter
relay router
free/—free/——free/—
routing.run
relay router
$0.100$0.100—$0.100*
DevPass (LLM Gateway)
relay router
$0.500$2.20$0.100$0.400
Kilo Gateway
relay router
$0.500$2.20$0.100$0.400
Eden AIoutlier
relay router
$0.500$2.20$0.100$0.400
Vercel AI Gateway
unknown
$0.600$2.40$0.120$0.458
Fireworks AI
relay router
$0.600$2.40$0.120$0.458
OpenRouter
unknown
$0.600$2.40$0.120$0.458
Pioneerbf16
relay router
$0.500$2.50$0.150$0.500
Nvidia
relay router
$0.500$2.50$0.150$0.459
NanoGPT
relay router
$0.500$2.50$0.250$0.523
Nebius Token Factoryoutlier
relay router
$1.00$3.00$1.00$1.18
Eden AIoutlier
relay router
$1.00$3.00$1.00$1.18
Together AIoutlier
open weights host
$0.600$3.60$0.200$0.618

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