Enterprise capability

Per-Project Routing Overrides

Different rules for different teams, in the same org

Most teams need one routing config for production, another for internal tools, and a third for experimentation. Per-Project Routing Overrides let you set defaults at the organization level and selectively override them on individual projects. Pin your billing-system project to a single US region for SOC 2 scope. Let your research project fall back to whatever provider is cheapest right now. Override fallback ordering, region locks, cost ceilings, and model allow-lists — all from a structured config that's auditable, diffable, and version-controlled.

Talk to salesTry LLM Gateway free

Why teams turn it on

Project-scoped config

Every project inherits the org default and can override any rule — region, provider priority, fallback chain, cost limits.

Region pinning

Lock a project to specific AWS / GCP / Azure regions for residency, latency, or compliance scope.

Per-project cost ceilings

Cap monthly spend per project. Auto-pause or auto-downgrade when a project's budget is hit — production is never starved by a runaway experiment.

Diffable + audited

Every override change is captured by [[audit-logs]] with a structured before/after diff. Roll back to any prior config in one click.

How it works

From decision to deployed in three short steps

  1. 01

    Set org-wide defaults

    Configure global routing rules — preferred providers, fallback order, region preferences, model allow-list.

  2. 02

    Override on the project

    In any project's Routing tab, override any subset of the org defaults. Inherited values stay live; overridden values take precedence.

  3. 03

    Inspect the effective config

    The Effective Config view shows the merged result for any project — what's inherited, what's overridden, and why.

json

Project routing override

{
  "projectId": "proj_billing_prod",
  "routing": {
    "regions": ["us-east-1"],
    "providers": ["aws-bedrock", "anthropic"],
    "fallback": "fail-fast",
    "monthlyBudgetUsd": 50000,
    "onBudgetExceeded": "pause"
  }
}

Real-world use cases

Why customers actually adopt this

01

SOC 2 scope reduction

Keep your audit-scope project pinned to us-east-1 with a single approved provider. Everything outside that scope is free to roam.

02

Multi-team isolation

Research, engineering, and customer-facing apps share an org but route independently — and bill independently.

03

Cost-aware experimentation

Cap experimental projects at $200/mo so a runaway agent loop can't burn the production budget overnight.

Frequently asked

Can a project completely diverge from the org default?
Yes. Every field is independently overridable. You can also lock specific fields at the org level to prevent project-level overrides for compliance reasons.
How do region pins interact with provider failover?
Region pins are honored during failover. If you pin a project to us-east-1, the fallback chain only considers providers/regions that match — never silently routes traffic to a different geography.

More enterprise capabilities

The rest of the enterprise stack

Organization-Wide Analytics

Cost, requests, and tokens totaled across every project, with breakdowns by model, project, API key, and member. Built on pre-aggregated rollups, so any date range stays fast.

Per-Member Budgets & Developer Role

Per-member spend caps enforced at request time, org-wide default developer limits, and a project-scoped Developer role. An over-budget request is rejected before it ever reaches a provider.

Enterprise Audit Logs

Tamper-evident audit trails for SOC 2, HIPAA, ISO 27001, and internal investigations. Every config change, key rotation, and admin action — captured, attributed, exportable.

Enterprise Guardrails

Server-side detection for prompt injection, PII, secrets, and policy violations. Configured centrally, enforced at the gateway, auditable per-request.

Discord & Slack Alerts

Native webhook integrations for Discord and Slack. Get the enterprise contact-sales form, billing events, guardrail trips, and SLA breaches in the channels your team already monitors.

Single Sign-On (SAML / OIDC)

SAML 2.0 and OIDC SSO with SCIM provisioning, group-based role mapping, and enforced-only access. No local credentials, no shared passkeys, no off-boarding gaps.

White-Label Chat & Playground

Embed or stand up a fully white-labeled chat app and playground under your own domain. Customize branding, default models, system prompts, and feature toggles.

Provider Compliance Policies

Define the certifications and data policies your providers must meet — SOC 2, ISO 27001, GDPR, no prompt training, no prompt logging — and the gateway refuses to route to anything that doesn't qualify.

See per-project routing overrides on your real workloads

Bring a sample workload to a 30-minute call. We'll wire it up live and show you the actual experience your team will get.

Book a walkthroughStart free