Nebius Token Factory (formerly Nebius AI Studio)
Nebius Group N.V. · nebius.com · Last checked:
- Serverless per-token API
- Dedicated endpoints
- Bring-your-own-model hosting
Company status
Active. Nebius is publicly listed (NASDAQ: NBIS). Token Factory is a business line of Nebius and the successor to Nebius AI Studio (launched November 5, 2025).
Sources 1, 2, 3, 4 · checked 2026-11-08
Capabilities
- Open-weight models (for example DeepSeek, GPT-OSS, Llama, Nemotron, Qwen) with OpenAI-compatible APIs
- Shared or single-tenant dedicated endpoints with a 99.9% SLA (vendor statement)
- Option to host your own models; fine-tuning and post-training
- Zero-retention inference in EU or US data centers for data-residency needs
Sources 1, 2, 3, 4 · checked 2026-11-08
Who it tends to fit
- Teams needing EU or US processing with zero retention for open-weight models
- Teams moving from a shared API to single-tenant endpoints with the same vendor
Sources 1, 2, 3, 4 · checked 2026-11-08
Pricing
Contact vendor or sign in: Nebius describes transparent $/token pricing, but its price list page asked for a login when checked on 2026-11-08.
Sources 1, 2, 3, 4 · checked 2026-11-08
Published certifications
Listed as stated by the vendor. Scope can differ by product or subsidiary; request current reports.
Sources 1, 2, 3, 4 · checked 2026-11-08
Limitations to check
- Price list required sign-in when checked
- Some serverless endpoints were flagged for deprecation; check the model you need
- Zero-retention terms and custom DPAs are on request; confirm in contract
Sources 1, 2, 3, 4 · checked 2026-11-08
Sources
- Enterprise inference page: dedicated or shared endpoints, zero-retention data flow, $/token pricing, 99.9% SLA, SOC 2 Type II, HIPAA, ISO 27001nebius.com/services/token-factory/enterprise-grade-inferenceChecked 2026-11-08
- Launch release: successor to AI Studio; zero-retention inference in EU or US datacenters; SOC 2 Type II including HIPAA, ISO 27001, ISO 27799; OpenAI-compatible APIs; host your own models; listed on Nasdaq (NBIS)nebius.com/newsroom/nebius-launches-nebius-token-factory-to-deliver-production-ai-inference-at-scaleChecked 2026-11-08
- Quickstart uses the OpenAI client with base URL api.tokenfactory.nebius.com/v1docs.tokenfactory.nebius.com/quickstartChecked 2026-11-08
- Price list URL redirects to a page that asks for login; banner on serverless endpoint deprecationsnebius.com/token-factory/pricesChecked 2026-11-08
Report a correction: [email protected]. Facts are re-checked at least every 90 days.