Vendors · Comparison
Databricks Mosaic AI Model Serving (now Databricks Model Serving) vs Nebius
Databricks Model Serving's parent is in the United States; Nebius's is in the Netherlands. Pricing, compliance and residency compared, with sources.
Last verified (older of the two profiles) How we verify
From the data
Key differences
- Pricing Databricks Model Serving: Per DBU per hour by GPU instance size (e.g. A10G 20 DBU/hour); per-token Foundation Model APIs (published pricing). Nebius: Per 1M input/output tokens; dedicated endpoints billable while a replica runs (unit not published) (publication not verified).
- Category Databricks Model Serving is a Managed self-host offering; Nebius is Dedicated GPU.
- Compliance Both document SOC 2 Type II, ISO 27001, HIPAA BAA and GDPR DPA. Only Databricks Model Serving documents FedRAMP. Only Nebius documents SOC 3.
- Data residency Databricks Model Serving states data residency on its own pages (not contractual); Nebius commits to data residency in its contract terms.
- Jurisdiction Parent jurisdiction: Databricks Model Serving United States, Nebius Netherlands.
Generated from the verified vendor data below; sources and dates on each vendor profile.
Databricks Mosaic AI Model Serving (now Databricks Model Serving)
Model Serving deploys custom MLflow models and fine-tuned foundation models on Databricks-managed serverless compute; provisioned throughput allocates dedicated inference capacity. Trust pages list ISO 27001:2022 (including AWS single-tenant), SOC 2 Type II, HIPAA options and FedRAMP Class C and D (formerly Moderate and High). A standard BAA and a DPA with SCCs are published.
Nebius
Token Factory offers shared or single-tenant endpoints with optional regional deployment in EU or US data centers (Finland, France, US). Non-US customers contract with Nebius B.V. under Dutch law; US Token Factory customers registering from 15 Sep 2026 contract with Nebius Inc. PHI use requires a signed BAA (HIPAA guideline). Per-token prices require sign-in.
Side by side
Capabilities compared
| Databricks Mosaic AI Model Serving (now Databricks Model Serving) | Nebius | |
|---|---|---|
| Pricing model | Per DBU per hour by GPU instance size (e.g. A10G 20 DBU/hour); per-token Foundation Model APIs | Per 1M input/output tokens; dedicated endpoints billable while a replica runs (unit not published) |
| Public pricing | Yes | Not verified |
| Category | Managed self-host | Dedicated GPU |
| Deployment models | Custom model servingProvisioned throughputPay-per-token endpoints | Serverless APIDedicated endpoints |
| Regions | United States, United Kingdom, EU/EEA, Asia-Pacific | EU/EEA, United States |
| Compliance | SOC 2 Type IIISO 27001HIPAA BAAFedRAMPGDPR DPA | SOC 2 Type II (stated)SOC 3 (stated)ISO 27001 (stated)GDPR DPAHIPAA BAA |
| Data residency evidence | Stated | Contractual |
| Parent jurisdiction | United States | Netherlands |
| Sovereign for | Not recorded | Not recorded |
| Analyst note | Model Serving deploys custom MLflow models and fine-tuned foundation models on Databricks-managed serverless compute; provisioned throughput allocates dedicated inference capacity. Trust pages list ISO 27001:2022 (including AWS single-tenant), SOC 2 Type II, HIPAA options and FedRAMP Class C and D (formerly Moderate and High). A standard BAA and a DPA with SCCs are published. | Token Factory offers shared or single-tenant endpoints with optional regional deployment in EU or US data centers (Finland, France, US). Non-US customers contract with Nebius B.V. under Dutch law; US Token Factory customers registering from 15 Sep 2026 contract with Nebius Inc. PHI use requires a signed BAA (HIPAA guideline). Per-token prices require sign-in. |
Where they diverge
Deployment differentiation
Only Databricks Model Serving
Both
No overlap in deployment models.
Only Nebius
Read the full profiles