Organization management

About

Each Prism organization is an isolated environment with its own providers, routing rules, rate limits, budgets, and API keys. Organizations are the top-level unit for multi-tenancy in Prism.

Organization settings

Organization config controls all gateway behavior for that org. Settings are managed via the dashboard or the admin API.

from prism import Prism

# base_url = inference gateway, control_plane_url = admin/config API
client = Prism(
    api_key="sk-prism-your-key",
    base_url="https://gateway.futureagi.com",
    control_plane_url="https://api.futureagi.com",
)

# Get org config
config = client.org_configs.retrieve(org_id="your-org-id")

# Update org config
client.org_configs.update(
    org_id="your-org-id",
    config={
        "rate_limiting": {
            "enabled": True,
            "rpm": 1000,
        },
        "budgets": {
            "limit": 500.00,
            "period": "monthly",
        },
    },
)

import { Prism } from "@futureagi/prism";

const client = new Prism({
    apiKey: "sk-prism-your-key",
    baseUrl: "https://gateway.futureagi.com",
    controlPlaneUrl: "https://api.futureagi.com",
});

const config = await client.orgConfigs.retrieve({
    orgId: "your-org-id",
});

await client.orgConfigs.update({
    orgId: "your-org-id",
    config: {
        rate_limiting: {
            enabled: true,
            rpm: 1000,
        },
        budgets: {
            limit: 500.0,
            period: "monthly",
        },
    },
});

Members and roles

Organizations can have multiple members with different roles.

Managing members

Members are managed through the Future AGI dashboard at Settings > Members. Invite new members by email. Each member can belong to multiple organizations.

API key management

Each organization has its own pool of API keys (virtual keys). Keys inherit org-level settings and can have additional per-key restrictions.

# List keys for an org
keys = client.keys.list(org_id="your-org-id")
for key in keys:
    print(f"{key.name}: {key.key_prefix}...")

# Create a new key
new_key = client.keys.create(
    org_id="your-org-id",
    name="backend-service",
    rate_limit_rpm=100,
    allowed_models=["gpt-4o", "gpt-4o-mini"],
)
print(f"Key: {new_key.key}")  # full key shown only at creation

# Revoke a key
client.keys.delete(key_id=new_key.id)

See Virtual keys & access control for detailed key configuration (RBAC, IP ACL, model restrictions).

Multi-tenancy patterns

One org per customer

For SaaS products, create a separate org per customer. Each customer gets isolated providers, budgets, and rate limits:

• Customer A: budget $100/month, access to gpt-4o-mini only
• Customer B: budget $500/month, access to gpt-4o and claude-sonnet-4-6
• Customer C: unlimited budget, all models

One org with per-key isolation

For internal teams, use a single org with per-key restrictions:

• Marketing team key: rate limit 50 RPM, budget $200/month
• Engineering team key: rate limit 500 RPM, budget $1000/month
• Data science key: rate limit 200 RPM, all models, no budget cap

Next Steps