inference/research
draftmedium · 0.70updated 2026-10-14

Nebius prices page and Token Factory (per-token inference vs GPU-hour split)

database
This record is a draft. Not yet verified. Provisional. Treat every number on this page as provisional.

Fields

id
sources/svc-src-nebius-token-factory-2026-10-14
Publisher
Nebius
Kind
database
Accessed
2026-10-14
notes
Fetched 2026-10-14. nebius.com/prices is GPU-hour infrastructure pricing (HGX H100 $4.50, H200 $5.40, B200 $8.50, B300 $9.50, RTX PRO 6000 $1.80; spot from $0.79) and explicitly links 'Token Factory prices' at tokenfactory.nebius.com for per-token managed inference - an authenticated console page this pass could not read. Draft confidence because the per-token rate card is second-hand; re-verify against the Token Factory page or the public docs mirror before quoting a number.

Notes

Fetched 2026-10-14. nebius.com/prices is GPU-hour infrastructure pricing (HGX H100 $4.50, H200 $5.40, B200 $8.50, B300 $9.50, RTX PRO 6000 $1.80; spot from $0.79) and explicitly links 'Token Factory prices' at tokenfactory.nebius.com for per-token managed inference - an authenticated console page this pass could not read. Draft confidence because the per-token rate card is second-hand; re-verify against the Token Factory page or the public docs mirror before quoting a number.

Cross-references

This record references no other records.