Scoped API keys & plans
Issue keys per workspace with chat, agent, or training scopes. Meter usage with abstract monthly token credits and optional trial limits—aligned with what you sell to customers.
Powered by Ollama · Multi-tenant · Usage you can bill against
Stop duct-taping auth, keys, and usage limits. SSIT AI Kit gives you scoped API keys, a self-serve portal, and clear monthly credits—so you can launch chat, agents, and training flows and grow into paid tiers with confidence.
Integrate once at /api/v1, issue keys per workspace, and let subscribers see usage in real time. Start free, prove value, then scale.
Built for teams who sell software—not for weekend scripts. Each block maps to something your roadmap already asks for.
Issue keys per workspace with chat, agent, or training scopes. Meter usage with abstract monthly token credits and optional trial limits—aligned with what you sell to customers.
Your users sign in, create keys, see usage, and upgrade paths. Operators can adjust catalog pricing and quotas; workspaces sync so limits always match published plans.
Admin tools for models, fine-tune jobs, registry, and billing hooks (e.g. Paymob). Integrate at /api/v1 with standard headers and documented routes.
Three straightforward steps. Your developers stay in flow; your ops stay in control.
Register in the customer portal. You receive a tenant with a plan, monthly credit bucket, and (on trial) a daily request cap on the welcome key.
Open Dashboard, create a key with the scope your plan allows, and call https://your-host/api/v1 with X-API-Key or Authorization: Bearer.
Use /chat, /generate, and optional /finetune flows. Monitor usage on the dashboard; upgrade plan or scopes when you need agents or training APIs.
Pick a tier that matches your stage. Upgrade when you need agents, training APIs, or higher monthly credits—catalog pricing stays in sync with your portal.
Chat API · bounded free tier for evaluation
Includes ~ 1,500 requests/mo via backend
Token credits / month: ~15,000
Billed monthly.
Scopes: chat
Higher chat limits for production apps
Includes ~ 120,000 requests/mo via backend
Token credits / month: ~600,000
Billed monthly.
Scopes: chat
Chat + create and manage agents
Includes ~ 300,000 requests/mo via backend
Token credits / month: ~1,500,000
Billed monthly.
Scopes: chat, chat_agent
Agents + fine-tuning and training APIs
Includes ~ 1,000,000 requests/mo via backend
Token credits / month: ~4,000,000
Billed monthly.
Scopes: chat, chat_agent, chat_agent_train
Your Ollama only — connect, fine-tune, and run models on hardware you control
Includes ~ 2,000,000 requests/mo via backend
Token credits / month: ~10,000,000
Billed monthly.
Scopes: chat, chat_agent, chat_agent_train, local_model_access
Custom limits and SLA — contact sales
Token credits: custom / negotiated
Scopes: chat, chat_agent, chat_agent_train
Abstract credits and scoped keys mean predictable margins and fewer surprises when models or traffic change.
Abstract credits per successful call simplify billing versus raw LLM tokens—easier to bundle into your own SaaS pricing.
Skip building auth, metering, and multi-tenancy from scratch; focus on product UX while SSIT AI Kit handles the API layer.
Progress from chat-only to agents and fine-tuned models as your plan and Ollama deployment allow—one consistent integration surface.
A multi-tenant API server for Ollama: scoped API keys, customer portal, and operator tools. You call /api/v1 with your key.
Create an account in the customer portal. You get one trial chat key with a daily request cap — no card required.
Sign in to the portal, open Dashboard, then create a key or claim the welcome trial key.
Use the host where this server runs, e.g. https://your-domain.com/api/v1 — after you create a key in the portal, use header X-API-Key.
They are a monthly usage budget on your plan—not raw LLM tokens. Each successful chat or generate call spends a small fixed number of credits (see your dashboard). When the bucket is used up, upgrade or wait for the next period.
SSIT is the engineering team behind this kit—book a call, learn about our company, or try our free site tools.
Quick answers about trials, API keys, and scopes…
A multi-tenant API server for Ollama: scoped API keys, customer portal, and operator tools. You call /api/v1 with your key.
Create an account in the customer portal. You get one trial chat key with a daily request cap — no card required.
Sign in to the portal, open Dashboard, then create a key or claim the welcome trial key.
Use the host where this server runs, e.g. https://your-domain.com/api/v1 — after you create a key in the portal, use header X-API-Key.
They are a monthly usage budget on your plan—not raw LLM tokens. Each successful chat or generate call spends a small fixed number of credits (see your dashboard). When the bucket is used up, upgrade or wait for the next period.