Which AI models and providers does Privian support?
Privian is a multi-provider LLM gateway. Through a single endpoint and request format, it currently routes to OpenAI, Anthropic, Google (Gemini), xAI, Mistral, Groq and DeepSeek. Every provider's models pass through the same gateway controls — including PII masking — before reaching the model.
Last reviewed: 2026-08-11 · Model and provider support below is derived directly from the live gateway model catalog, not a separately maintained list.
How model support works
- Provider-namespaced IDs. Every model is addressed as
provider/model(e.g.anthropic/claude-sonnet-4-5), so you can switch providers without changing integration code. - BYOK credentials. Bring your own provider API keys and keep billing and rate limits on your existing provider accounts.
- Masking applies uniformly. PII masking and other gateway policies apply the same way regardless of which provider serves the request.
Learn more: LLM Gateway · PII masking · Pricing · Docs
Loading models…
Use the model ID directly in your API requests — Privian never rewrites or aliases the model you select.
Frequently asked questions
- Which providers does Privian support?
- Privian routes to multiple model providers — including OpenAI, Anthropic, Google (Gemini), xAI, Mistral, Groq and DeepSeek — through a single gateway endpoint. The exact list of live models is published on this page and sourced from the gateway model catalog.
- Can I use OpenAI and Anthropic through one endpoint?
- Yes. Privian exposes one API endpoint and one request format; you select the provider and model via a provider-namespaced model ID (e.g. openai/gpt-4o-mini or anthropic/claude-sonnet-4-5) rather than integrating each provider's SDK separately.
- How are models addressed?
- Every model is addressed by a provider-namespaced ID in the form provider/model, such as openai/gpt-5.5 or google/gemini-2.5-pro. Privian does not rewrite or alias the model you select.
- Do I need my own provider keys?
- Privian supports bring-your-own-key (BYOK): you can connect your own provider credentials so calls are billed and rate-limited under your existing provider accounts, while still passing through Privian's masking and gateway controls.
- What happens when a provider adds a new model?
- New models are added to the gateway catalog as they become available and appear in this list with their status (active, preview, or deprecated). Masking and gateway controls apply automatically to any newly added model, regardless of provider.