# Agentic Router — One API Key for Every Model, Every Tool and Every Agent You Own > Source: https://ibl.ai/product/agentic-router One API key for every model and every tool: 45 providers and 316 chat models behind one endpoint, 1,000+ MCP servers, and your own agents callable from REST, WebSocket, the SDK, agent skills or any MCP client, with unified billing and spend caps, on infrastructure you own. On ibl.ai you own all the code and the data, run it model-agnostic across any LLM, and pay with no per-seat pricing — so you can deploy anywhere, from your own cloud to a fully air-gapped network. **One API key for every model and tool** **1.6M+** Platform Users · **400+** Organizations · Trusted by Syracuse University, SUNY, Google, NVIDIA, Microsoft, OpenAI and Kaplan. > “ibl.ai does amazing work and are great people.” — Paul LeBlanc, former President and CEO, Southern New Hampshire University Don’t take our word for it? [See what our partners say](https://ibl.ai/case-studies) [Get Started Free](https://ibl.ai/join) · [Read the skills](https://ibl.ai/developer/skills) · [Book a Demo](https://cal.com/iblai/30min) · [Contact Us](https://ibl.ai/contact) ## Two ways to run it ### We host it: Start at $0 on the ibl.ai cloud Free tier with renewable monthly credits, no credit card, usage-based from there. Move to your own infrastructure anytime, nothing to rebuild. [Hosted deployment](https://ibl.ai/hosted-deployment) · [See pricing](https://ibl.ai/pricing) ### You own it: Take the whole codebase Backend and infrastructure under a perpetual license, so you own all the code and the data. Your cloud, on-premise or air-gapped, under your control. [Full code license](https://ibl.ai/full-code-license) ## Available on - **Web** — Any modern browser, nothing to install ([details](https://ibl.ai/download#platforms)) - **Chrome** — Side-panel extension: your agents on any page ([details](https://ibl.ai/download#platforms)) - **macOS** — Universal desktop app, signed and notarized ([details](https://ibl.ai/download#platforms)) - **Windows** — Desktop installer for x64 and ARM64 ([details](https://ibl.ai/download#platforms)) - **iOS** — iPhone and iPad on the App Store ([details](https://ibl.ai/download#platforms)) - **Android** — Phones and tablets on Google Play ([details](https://ibl.ai/download#platforms)) - **Linux** — Build from source on GitHub ([details](https://ibl.ai/download#platforms)) ## What is behind the key Every model the platform serves, every tool reachable over MCP, and every agent your organization has built, from REST, WebSocket, the SDK, agent skills or any MCP client, on one bill with ceilings that hold. - **[Every model, one endpoint](https://ibl.ai/updates/llm-catalog-45-providers-any-model)** — 45 providers and 316 chat models on one grid, through OpenRouter, Vertex, Foundry, Bedrock and each provider's own API. Switch a model per request or per agent; prompts, data and connectors stay. The catalogue syncs itself behind a verification gate. (45 providers · 316 models) - **[Every tool, over MCP](https://ibl.ai/developer/platform/mcp/mcp-connections)** — Register any of the 1,000+ public MCP servers or your own, attach it to an agent, and bind credentials for the organization, one agent or one user. Tools read your systems in place; no copy of the data. (1,000+ MCP servers) - **[Your agents, two ways to authenticate](https://ibl.ai/developer)** — Call any agent you built from your own code with the user's own token and role, or server to server with a tenant-scoped platform API key. One bill, spend caps at three scopes, cost per call. (REST · WebSocket · SDK · skills · MCP) ## How is Agentic Router different from a model gateway? A model gateway puts one key in front of several model vendors and bills you per token. Agentic Router puts one key in front of the models, the tools and the agents you own, authenticates people and services differently, and runs where your data already is. | | Agentic Router | A hosted model gateway | |---|---|---| | **What the key reaches** | Every model, every MCP tool, and every agent you built. | Models only. | | **Authentication** | Per-user tokens with roles enforced, or a tenant-scoped platform key for services. | One organization key for everyone. | | **Billing and caps** | One bill across models and tools; caps per organization, agent and user; cost per call. | Per-model spend, one cap if any. | | **Keys and routing** | Your provider keys first; the highest-priority gateway that can serve the model. | The vendor's keys, the vendor's routing. | | **Where it runs** | Your cloud, on-premise or air-gapped, with models served inside the perimeter. | The vendor's cloud; every prompt passes through it. | | **What you own** | The code and the data. | A subscription. | ## Why ibl.ai - **Your Code. Your Data.** — Full ownership, zero vendor lock-in ([more](https://ibl.ai/full-code-license)) - **Any LLM, Your Choice** — Air-gapped or any cloud provider’s ([more](https://ibl.ai/self-hosted-ai)) - **85% Lower Cost** — Significantly cheaper at scale ([more](https://ibl.ai/ai-cost-calculator)) - **Engineers Included** — Our team builds it with yours ([more](https://ibl.ai/service/ai-transformation)) Agentic Router by ibl.ai is the engineering door to the platform: one API key that reaches 45 providers and 316 chat models, 1,000+ MCP servers and the agents your organization has built, from REST, WebSocket, the SDK, agent skills or any MCP client. Authenticate per user, with each person's own token and role, or server to server with a tenant-scoped platform key. One bill, spend caps at three scopes, cost per call. Deployed in your infrastructure with your keys and your controls—including full code access. ## For Developers and platform teams: One Key, Every Model and Tool ### Every model, one endpoint 45 providers and 316 chat models on one grid as of November 6, 2026: Anthropic, Google, OpenAI, Meta, Mistral, DeepSeek, Qwen, xAI and the open-weight labs, through OpenRouter, Vertex, Foundry, Bedrock and each provider's own API. Switch a model per request or per agent; prompts, data and connectors stay. ### Every tool, over MCP Register any of the 1,000+ public MCP servers or your own, attach it to an agent, and bind credentials at the scope you choose: the whole organization, one agent or one user. The tools reach your systems of record, files and SaaS tools in place, with no copy of the data. ### Your own agents, from your own code Every agent your organization has built on Agentic OS is callable from your apps, portals and pipelines: a REST call, a streaming WebSocket session, the SDK, an agent skill or any MCP client. For example: "Put an agent you built behind your own app, with each user's own token and role." ### Result One integration instead of a vendor account per model and a connector per tool. ## For Platform Teams: One Bill, Enforced Ceilings ### Unified billing Every model and every tool behind the key lands on one bill, priced by usage, never by seat. The Cost tab breaks spend down by provider, model, user and individual call, with latency percentiles beside it. ### Spend caps at three scopes A ceiling for the whole organization, for one agent, and for one person on one agent, each with its own reset period and a mode that blocks or alerts at the limit. Enforcement is server-side, so a runaway script stops at the cap. ### Your keys first A request routes to the highest-priority gateway that can serve the model and has a usable key, and your organization's own provider keys are preferred over platform keys. A deployment with provider keys and no ibl.ai key works with no extra configuration. ### A catalogue that keeps itself current The model list syncs every six hours and a new model goes live only after a verification gate has tested multi-turn recall, streaming, tool calling, image input and a real, non-zero cost. For example: "Route one task to a commodity model and another to a frontier one, from the same key." ### Result Cost you can see per call and cap per scope, with no procurement cycle per model. ## For Security & IT: Two Ways to Authenticate, One Place to Audit ### Per user People sign in with the identity they already have, and every call carries that person's own token. Roles are enforced on each request, so an agent answers only what the caller may see, and a person's in-chat model choice stays inside the list you allow. ### Server to server A tenant-scoped platform API key, sent as an Api-Token header, authenticates your backend: mint per-user tokens, provision users, run pipelines. It can act for any user on the tenant, so it lives on your server, never in browser code. ### Where it runs Your cloud, on-premise or fully air-gapped, where the same grid points at models served inside the perimeter. For example: "Give a pipeline a platform key with a monthly cap that blocks at the limit." ### Every call logged Who asked, which model answered, which tools were called and what it cost, on systems you administer, with replayable transcripts and an audit trail per agent. ### Result One key policy for people and for services, and nothing to reconcile across vendors. ## How Do You Call Agentic Router? ### REST More than 1,000 documented endpoints: agents, sessions, datasets, memory, MCP servers and connections, users, roles, analytics and billing. Authorization: Api-Token for a service, or the user's own token. ### Streaming Chat sessions stream over WebSocket, so a response renders token by token in your own interface. ### SDK and CLI @iblai/iblai-js for web and native apps, @iblai/cli for the terminal, and a Python quickstart (iblai/quickstarts) that creates an agent and chats with it from a script. ### Agent skills 161 skills from iblai/vibe teach Claude Code, Codex, Cursor, OpenCode or any skills-compatible coding agent to run the platform itself: configure agents, manage datasets and memory, pull cost analytics, each mapped to exact REST calls. Every skill is documented at ibl.ai/developer/skills; install with npx -y @iblai/mcp. ### Any MCP client Every agent you deploy can be added to Claude Code, Cursor, Claude Desktop or any other MCP client in one step, and talked to from there with its own knowledge, memory and guardrails. ### Embed and proxy Embed the chat widget in your own app with tokens provisioned from your backend, no login popup. Third-party services such as ElevenLabs and HeyGen sit behind the same key through the external service proxy. ## What's Inside Agentic Router (Highlights) ### Model catalogue 45 providers, 316 chat models, synced every six hours behind a verification gate ### Gateways OpenRouter, Vertex, Foundry, Bedrock and each provider's own API, your keys preferred ### MCP servers 1,000+ public servers plus your own; credentials bound per organization, agent or user; OAuth-backed connectors ### Your agents Every agent on Agentic OS callable by ID, with its knowledge, memory and guardrails ### Two auth modes Per-user tokens with roles enforced, or a tenant-scoped platform API key for services ### Spend caps Organization, agent and per-user ceilings; block or alert; reset periods ### Cost per call Spend by provider, model, user and call, with latency percentiles ### Surfaces REST, WebSocket, @iblai/iblai-js, @iblai/cli, 161 agent skills for Claude Code, Codex and Cursor, any MCP client, embed, external service proxy ## Why Agentic Router Instead of a Model Gateway? ### What the key reaches Agentic Router: Every model, every MCP tool and every agent you own | Model gateway: Models only ### Authentication Agentic Router: Per-user tokens with roles, or a tenant-scoped platform key | Model gateway: One organization key for everyone ### Billing Agentic Router: One bill across models and tools, caps per organization, agent and user | Model gateway: Per-model spend, one cap if any ### Where it runs Agentic Router: Your cloud, on-premise or air-gapped, your provider keys first | Model gateway: The vendor's cloud, through the vendor's keys ### Your data Agentic Router: Read in place over MCP, scoped to the caller, every call logged on your systems | Model gateway: Every prompt passes through a third party ### Ownership Agentic Router: You own all the code and the data | Model gateway: A subscription ## Benefits That Drive Impact ### Developers and platform teams One key and one SDK instead of a vendor account per model and a connector per tool. ### Platform teams One bill, cost per call, and ceilings that hold without a meeting. ### Security and leadership Two auth modes under one policy, every call logged, and the whole stack owned rather than rented. ## Part of the ibl.ai platform Agentic Router is how engineers reach the platform: Agentic OS runs the agents, Agentic Ontology exposes your systems over MCP, Agentic Vibe builds the apps that call them, and Agentic LMS, SIS, Wallet and CRM are applications built the same way. One key, one bill, one platform you own. ## Experience Agentic Router Today ### Start for Free Start at ibl.ai/join, open an agent in Agentic OS, and create a platform API key on its API tab ### Read the skills 161 agent skills documented at ibl.ai/developer/skills, one per capability, each mapped to exact REST calls, for Claude Code, Codex, Cursor and any skills-compatible agent ### Read the API reference Every endpoint, live from the platform's OpenAPI document, at ibl.ai/developer/api-reference ### Schedule a Demo Walk through your integration, your models and your cost controls with our engineers ## By Sector This page has a sector selector. Every sector is listed below with its own URL; where a sector's body differs from the one above, it is reproduced in full — that is the content a visitor sees after choosing that sector. ### Higher Education > Source: https://ibl.ai/product/agentic-router/higher-education One API key for a university's developers: every model, 1,000+ MCP servers and the institution's own agents, callable from the portal, the LMS or a script, per student or server to server, with one bill and spend caps. On ibl.ai you own all the code and the data, run it model-agnostic across any LLM, and pay with no per-seat pricing — so you can deploy anywhere, from your own cloud to a fully air-gapped network. Agentic Router by ibl.ai gives an institution's developers one key to everything the platform serves: 45 providers and 316 chat models, the MCP tools that read the SIS and LMS in place, and the agents the campus has already built. Call it from the student portal, the LMS, a research pipeline or a nightly job, with each student's own token or a platform key, and see what every call cost. Deployed in your infrastructure with your keys and your controls—including full code access. #### For The IT and portal teams: One Key, Every Model and Tool ##### Every model, one endpoint 45 providers and 316 chat models on one grid as of November 6, 2026: Anthropic, Google, OpenAI, Meta, Mistral, DeepSeek, Qwen, xAI and the open-weight labs, through OpenRouter, Vertex, Foundry, Bedrock and each provider's own API. Switch a model per request or per agent; prompts, data and connectors stay. ##### Every tool, over MCP Register any of the 1,000+ public MCP servers or your own, attach it to an agent, and bind credentials at the scope you choose: the whole organization, one agent or one user. The tools reach Banner or PeopleSoft, Canvas and Salesforce in place, with no copy of the data. ##### Your own agents, from your own code Every agent your organization has built on Agentic OS is callable from the student portal, the LMS and the advising app: a REST call, a streaming WebSocket session, the SDK, an agent skill or any MCP client. For example: "Add the advising agent to the student portal, with each student's own token and role." ##### Result One integration instead of a vendor account per model and a connector per tool. #### For Platform Teams: One Bill, Enforced Ceilings ##### Unified billing Every model and every tool behind the key lands on one bill, priced by usage, never by seat. The Cost tab breaks spend down by provider, model, user and individual call, with latency percentiles beside it. ##### Spend caps at three scopes A ceiling for the whole organization, for one agent, and for one person on one agent, each with its own reset period and a mode that blocks or alerts at the limit. Enforcement is server-side, so a runaway script stops at the cap. ##### Your keys first A request routes to the highest-priority gateway that can serve the model and has a usable key, and your organization's own provider keys are preferred over platform keys. A deployment with provider keys and no ibl.ai key works with no extra configuration. ##### A catalogue that keeps itself current The model list syncs every six hours and a new model goes live only after a verification gate has tested multi-turn recall, streaming, tool calling, image input and a real, non-zero cost. For example: "Route a syllabus draft to the cheapest model that passes your test set, and switch when a better one ships." ##### Result Cost you can see per call and cap per scope, with no procurement cycle per model. #### For Security & IT: Two Ways to Authenticate, One Place to Audit ##### Per user Students, faculty and staff sign in with the identity they already have, and every call carries that person's own token. Roles are enforced on each request, so an agent answers only what the caller may see, and a person's in-chat model choice stays inside the list you allow. ##### Server to server A tenant-scoped platform API key, sent as an Api-Token header, authenticates your backend: mint per-user tokens, provision users, run pipelines. It can act for any user on the tenant, so it lives on your server, never in browser code. ##### Where it runs Your cloud, on-premise or fully air-gapped, where the same grid points at models served inside the perimeter. For example: "Expose the SIS over MCP to an agent, read-only, scoped to the caller." ##### Every call logged Who asked, which model answered, which tools were called and what it cost, on systems you administer, with replayable transcripts and an audit trail per agent. ##### Result One key policy for people and for services, and nothing to reconcile across vendors. #### How Do You Call Agentic Router? ##### REST More than 1,000 documented endpoints: agents, sessions, datasets, memory, MCP servers and connections, users, roles, analytics and billing. Authorization: Api-Token for a service, or the user's own token. ##### Streaming Chat sessions stream over WebSocket, so a response renders token by token in your own interface. ##### SDK and CLI @iblai/iblai-js for web and native apps, @iblai/cli for the terminal, and a Python quickstart (iblai/quickstarts) that creates an agent and chats with it from a script. ##### Agent skills 161 skills from iblai/vibe teach Claude Code, Codex, Cursor, OpenCode or any skills-compatible coding agent to run the platform itself: configure agents, manage datasets and memory, pull cost analytics, each mapped to exact REST calls. Every skill is documented at ibl.ai/developer/skills; install with npx -y @iblai/mcp. ##### Any MCP client Every agent you deploy can be added to Claude Code, Cursor, Claude Desktop or any other MCP client in one step, and talked to from there with its own knowledge, memory and guardrails. ##### Embed and proxy Embed the chat widget in your own app with tokens provisioned from your backend, no login popup. Third-party services such as ElevenLabs and HeyGen sit behind the same key through the external service proxy. #### What's Inside Agentic Router (Highlights) ##### Model catalogue 45 providers, 316 chat models, synced every six hours behind a verification gate ##### Gateways OpenRouter, Vertex, Foundry, Bedrock and each provider's own API, your keys preferred ##### MCP servers 1,000+ public servers plus your own; credentials bound per organization, agent or user; OAuth-backed connectors ##### Your agents Every agent on Agentic OS callable by ID, with its knowledge, memory and guardrails ##### Two auth modes Per-user tokens with roles enforced, or a tenant-scoped platform API key for services ##### Spend caps Organization, agent and per-user ceilings; block or alert; reset periods ##### Cost per call Spend by provider, model, user and call, with latency percentiles ##### Surfaces REST, WebSocket, @iblai/iblai-js, @iblai/cli, 161 agent skills for Claude Code, Codex and Cursor, any MCP client, embed, external service proxy #### Why Agentic Router Instead of a Model Gateway? ##### What the key reaches Agentic Router: Every model, every MCP tool and every agent you own | Model gateway: Models only ##### Authentication Agentic Router: Per-user tokens with roles, or a tenant-scoped platform key | Model gateway: One organization key for everyone ##### Billing Agentic Router: One bill across models and tools, caps per organization, agent and user | Model gateway: Per-model spend, one cap if any ##### Where it runs Agentic Router: Your cloud, on-premise or air-gapped, your provider keys first | Model gateway: The vendor's cloud, through the vendor's keys ##### Your data Agentic Router: Read in place over MCP, scoped to the caller, every call logged on your systems | Model gateway: Every prompt passes through a third party ##### Ownership Agentic Router: You own all the code and the data | Model gateway: A subscription #### Benefits That Drive Impact ##### The IT and portal teams One key and one SDK instead of a vendor account per model and a connector per tool. ##### Platform teams One bill, cost per call, and ceilings that hold without a meeting. ##### Security and leadership Two auth modes under one policy, every call logged, and the whole stack owned rather than rented. #### Part of the ibl.ai platform Agentic Router is how engineers reach the platform: Agentic OS runs the agents, Agentic Ontology exposes your systems over MCP, Agentic Vibe builds the apps that call them, and Agentic LMS, SIS, Wallet and CRM are applications built the same way. One key, one bill, one platform you own. #### Experience Agentic Router Today ##### Start for Free Start at ibl.ai/join, open an agent in Agentic OS, and create a platform API key on its API tab ##### Read the skills 161 agent skills documented at ibl.ai/developer/skills, one per capability, each mapped to exact REST calls, for Claude Code, Codex, Cursor and any skills-compatible agent ##### Read the API reference Every endpoint, live from the platform's OpenAPI document, at ibl.ai/developer/api-reference ##### Schedule a Demo Walk through your integration, your models and your cost controls with our engineers ### Corporate > Source: https://ibl.ai/product/agentic-router/corporate One API key for the enterprise's platform team: every model, 1,000+ MCP servers and your own agents behind one endpoint, per employee or server to server, with unified billing, spend caps and cost per call. On ibl.ai you own all the code and the data, run it model-agnostic across any LLM, and pay with no per-seat pricing — so you can deploy anywhere, from your own cloud to a fully air-gapped network. Agentic Router by ibl.ai gives a platform team one key to everything the platform serves: 45 providers and 316 chat models, the MCP tools that reach Salesforce, ServiceNow and Workday in place, and the agents the company has already built. Call it from the intranet, the support desk or a pipeline, with each employee's own token or a platform key, and hold the ceiling with spend caps at three scopes. Deployed in your infrastructure with your keys and your controls—including full code access. #### For Platform and integration teams: One Key, Every Model and Tool ##### Every model, one endpoint 45 providers and 316 chat models on one grid as of November 6, 2026: Anthropic, Google, OpenAI, Meta, Mistral, DeepSeek, Qwen, xAI and the open-weight labs, through OpenRouter, Vertex, Foundry, Bedrock and each provider's own API. Switch a model per request or per agent; prompts, data and connectors stay. ##### Every tool, over MCP Register any of the 1,000+ public MCP servers or your own, attach it to an agent, and bind credentials at the scope you choose: the whole organization, one agent or one user. The tools reach Salesforce, ServiceNow, Workday and SharePoint in place, with no copy of the data. ##### Your own agents, from your own code Every agent your organization has built on Agentic OS is callable from the intranet, the support desk and internal tools: a REST call, a streaming WebSocket session, the SDK, an agent skill or any MCP client. For example: "Put the IT help-desk agent in Slack and ServiceNow, with each employee's own token." ##### Result One integration instead of a vendor account per model and a connector per tool. #### For Platform Teams: One Bill, Enforced Ceilings ##### Unified billing Every model and every tool behind the key lands on one bill, priced by usage, never by seat. The Cost tab breaks spend down by provider, model, user and individual call, with latency percentiles beside it. ##### Spend caps at three scopes A ceiling for the whole organization, for one agent, and for one person on one agent, each with its own reset period and a mode that blocks or alerts at the limit. Enforcement is server-side, so a runaway script stops at the cap. ##### Your keys first A request routes to the highest-priority gateway that can serve the model and has a usable key, and your organization's own provider keys are preferred over platform keys. A deployment with provider keys and no ibl.ai key works with no extra configuration. ##### A catalogue that keeps itself current The model list syncs every six hours and a new model goes live only after a verification gate has tested multi-turn recall, streaming, tool calling, image input and a real, non-zero cost. For example: "Send summarization to a commodity model and contract review to a frontier one, from the same key." ##### Result Cost you can see per call and cap per scope, with no procurement cycle per model. #### For Security & IT: Two Ways to Authenticate, One Place to Audit ##### Per user Employees and customers sign in with the identity they already have, and every call carries that person's own token. Roles are enforced on each request, so an agent answers only what the caller may see, and a person's in-chat model choice stays inside the list you allow. ##### Server to server A tenant-scoped platform API key, sent as an Api-Token header, authenticates your backend: mint per-user tokens, provision users, run pipelines. It can act for any user on the tenant, so it lives on your server, never in browser code. ##### Where it runs Your cloud, on-premise or fully air-gapped, where the same grid points at models served inside the perimeter. For example: "Give a data pipeline a platform key with a monthly cap that blocks at the limit." ##### Every call logged Who asked, which model answered, which tools were called and what it cost, on systems you administer, with replayable transcripts and an audit trail per agent. ##### Result One key policy for people and for services, and nothing to reconcile across vendors. #### How Do You Call Agentic Router? ##### REST More than 1,000 documented endpoints: agents, sessions, datasets, memory, MCP servers and connections, users, roles, analytics and billing. Authorization: Api-Token for a service, or the user's own token. ##### Streaming Chat sessions stream over WebSocket, so a response renders token by token in your own interface. ##### SDK and CLI @iblai/iblai-js for web and native apps, @iblai/cli for the terminal, and a Python quickstart (iblai/quickstarts) that creates an agent and chats with it from a script. ##### Agent skills 161 skills from iblai/vibe teach Claude Code, Codex, Cursor, OpenCode or any skills-compatible coding agent to run the platform itself: configure agents, manage datasets and memory, pull cost analytics, each mapped to exact REST calls. Every skill is documented at ibl.ai/developer/skills; install with npx -y @iblai/mcp. ##### Any MCP client Every agent you deploy can be added to Claude Code, Cursor, Claude Desktop or any other MCP client in one step, and talked to from there with its own knowledge, memory and guardrails. ##### Embed and proxy Embed the chat widget in your own app with tokens provisioned from your backend, no login popup. Third-party services such as ElevenLabs and HeyGen sit behind the same key through the external service proxy. #### What's Inside Agentic Router (Highlights) ##### Model catalogue 45 providers, 316 chat models, synced every six hours behind a verification gate ##### Gateways OpenRouter, Vertex, Foundry, Bedrock and each provider's own API, your keys preferred ##### MCP servers 1,000+ public servers plus your own; credentials bound per organization, agent or user; OAuth-backed connectors ##### Your agents Every agent on Agentic OS callable by ID, with its knowledge, memory and guardrails ##### Two auth modes Per-user tokens with roles enforced, or a tenant-scoped platform API key for services ##### Spend caps Organization, agent and per-user ceilings; block or alert; reset periods ##### Cost per call Spend by provider, model, user and call, with latency percentiles ##### Surfaces REST, WebSocket, @iblai/iblai-js, @iblai/cli, 161 agent skills for Claude Code, Codex and Cursor, any MCP client, embed, external service proxy #### Why Agentic Router Instead of a Model Gateway? ##### What the key reaches Agentic Router: Every model, every MCP tool and every agent you own | Model gateway: Models only ##### Authentication Agentic Router: Per-user tokens with roles, or a tenant-scoped platform key | Model gateway: One organization key for everyone ##### Billing Agentic Router: One bill across models and tools, caps per organization, agent and user | Model gateway: Per-model spend, one cap if any ##### Where it runs Agentic Router: Your cloud, on-premise or air-gapped, your provider keys first | Model gateway: The vendor's cloud, through the vendor's keys ##### Your data Agentic Router: Read in place over MCP, scoped to the caller, every call logged on your systems | Model gateway: Every prompt passes through a third party ##### Ownership Agentic Router: You own all the code and the data | Model gateway: A subscription #### Benefits That Drive Impact ##### Platform and integration teams One key and one SDK instead of a vendor account per model and a connector per tool. ##### Platform teams One bill, cost per call, and ceilings that hold without a meeting. ##### Security and leadership Two auth modes under one policy, every call logged, and the whole stack owned rather than rented. #### Part of the ibl.ai platform Agentic Router is how engineers reach the platform: Agentic OS runs the agents, Agentic Ontology exposes your systems over MCP, Agentic Vibe builds the apps that call them, and Agentic LMS, SIS, Wallet and CRM are applications built the same way. One key, one bill, one platform you own. #### Experience Agentic Router Today ##### Start for Free Start at ibl.ai/join, open an agent in Agentic OS, and create a platform API key on its API tab ##### Read the skills 161 agent skills documented at ibl.ai/developer/skills, one per capability, each mapped to exact REST calls, for Claude Code, Codex, Cursor and any skills-compatible agent ##### Read the API reference Every endpoint, live from the platform's OpenAPI document, at ibl.ai/developer/api-reference ##### Schedule a Demo Walk through your integration, your models and your cost controls with our engineers ### Small Business > Source: https://ibl.ai/product/agentic-router/small-business One API key for the developer or agency a small business works with: every model, 1,000+ MCP servers and your agents from your website or back office, with one bill and a cap you set. On ibl.ai you own all the code and the data, run it model-agnostic across any LLM, and pay with no per-seat pricing — so you can deploy anywhere, from your own cloud to a fully air-gapped network. Agentic Router by ibl.ai gives the developer you work with one key to everything the platform serves: dozens of model providers, the MCP tools that reach your accounting, calendar and email, and the agents you have already set up. Call it from your website, your booking flow or a spreadsheet script, with one bill and a monthly cap, instead of five vendor accounts. Hosted by ibl.ai or deployed on your own infrastructure—including full code access. #### For The developer or agency you work with: One Key, Every Model and Tool ##### Every model, one endpoint 45 providers and 316 chat models on one grid as of November 6, 2026: Anthropic, Google, OpenAI, Meta, Mistral, DeepSeek, Qwen, xAI and the open-weight labs, through OpenRouter, Vertex, Foundry, Bedrock and each provider's own API. Switch a model per request or per agent; prompts, data and connectors stay. ##### Every tool, over MCP Register any of the 1,000+ public MCP servers or your own, attach it to an agent, and bind credentials at the scope you choose: the whole organization, one agent or one user. The tools reach QuickBooks, your calendar and email in place, with no copy of the data. ##### Your own agents, from your own code Every agent your organization has built on Agentic OS is callable from your website, your booking flow and your back office: a REST call, a streaming WebSocket session, the SDK, an agent skill or any MCP client. For example: "Put the quote agent behind the contact form on your website." ##### Result One integration instead of a vendor account per model and a connector per tool. #### For Platform Teams: One Bill, Enforced Ceilings ##### Unified billing Every model and every tool behind the key lands on one bill, priced by usage, never by seat. The Cost tab breaks spend down by provider, model, user and individual call, with latency percentiles beside it. ##### Spend caps at three scopes A ceiling for the whole organization, for one agent, and for one person on one agent, each with its own reset period and a mode that blocks or alerts at the limit. Enforcement is server-side, so a runaway script stops at the cap. ##### Your keys first A request routes to the highest-priority gateway that can serve the model and has a usable key, and your organization's own provider keys are preferred over platform keys. A deployment with provider keys and no ibl.ai key works with no extra configuration. ##### A catalogue that keeps itself current The model list syncs every six hours and a new model goes live only after a verification gate has tested multi-turn recall, streaming, tool calling, image input and a real, non-zero cost. For example: "Let a spreadsheet script ask the bookkeeping agent a question each morning." ##### Result Cost you can see per call and cap per scope, with no procurement cycle per model. #### For Security & IT: Two Ways to Authenticate, One Place to Audit ##### Per user Your staff and your customers sign in with the identity they already have, and every call carries that person's own token. Roles are enforced on each request, so an agent answers only what the caller may see, and a person's in-chat model choice stays inside the list you allow. ##### Server to server A tenant-scoped platform API key, sent as an Api-Token header, authenticates your backend: mint per-user tokens, provision users, run pipelines. It can act for any user on the tenant, so it lives on your server, never in browser code. ##### Where it runs Your cloud, on-premise or fully air-gapped, where the same grid points at models served inside the perimeter. For example: "Set a monthly cap so the bill never surprises you." ##### Every call logged Who asked, which model answered, which tools were called and what it cost, on systems you administer, with replayable transcripts and an audit trail per agent. ##### Result One key policy for people and for services, and nothing to reconcile across vendors. #### How Do You Call Agentic Router? ##### REST More than 1,000 documented endpoints: agents, sessions, datasets, memory, MCP servers and connections, users, roles, analytics and billing. Authorization: Api-Token for a service, or the user's own token. ##### Streaming Chat sessions stream over WebSocket, so a response renders token by token in your own interface. ##### SDK and CLI @iblai/iblai-js for web and native apps, @iblai/cli for the terminal, and a Python quickstart (iblai/quickstarts) that creates an agent and chats with it from a script. ##### Agent skills 161 skills from iblai/vibe teach Claude Code, Codex, Cursor, OpenCode or any skills-compatible coding agent to run the platform itself: configure agents, manage datasets and memory, pull cost analytics, each mapped to exact REST calls. Every skill is documented at ibl.ai/developer/skills; install with npx -y @iblai/mcp. ##### Any MCP client Every agent you deploy can be added to Claude Code, Cursor, Claude Desktop or any other MCP client in one step, and talked to from there with its own knowledge, memory and guardrails. ##### Embed and proxy Embed the chat widget in your own app with tokens provisioned from your backend, no login popup. Third-party services such as ElevenLabs and HeyGen sit behind the same key through the external service proxy. #### What's Inside Agentic Router (Highlights) ##### Model catalogue 45 providers, 316 chat models, synced every six hours behind a verification gate ##### Gateways OpenRouter, Vertex, Foundry, Bedrock and each provider's own API, your keys preferred ##### MCP servers 1,000+ public servers plus your own; credentials bound per organization, agent or user; OAuth-backed connectors ##### Your agents Every agent on Agentic OS callable by ID, with its knowledge, memory and guardrails ##### Two auth modes Per-user tokens with roles enforced, or a tenant-scoped platform API key for services ##### Spend caps Organization, agent and per-user ceilings; block or alert; reset periods ##### Cost per call Spend by provider, model, user and call, with latency percentiles ##### Surfaces REST, WebSocket, @iblai/iblai-js, @iblai/cli, 161 agent skills for Claude Code, Codex and Cursor, any MCP client, embed, external service proxy #### Why Agentic Router Instead of a Model Gateway? ##### What the key reaches Agentic Router: Every model, every MCP tool and every agent you own | Model gateway: Models only ##### Authentication Agentic Router: Per-user tokens with roles, or a tenant-scoped platform key | Model gateway: One organization key for everyone ##### Billing Agentic Router: One bill across models and tools, caps per organization, agent and user | Model gateway: Per-model spend, one cap if any ##### Where it runs Agentic Router: Your cloud, on-premise or air-gapped, your provider keys first | Model gateway: The vendor's cloud, through the vendor's keys ##### Your data Agentic Router: Read in place over MCP, scoped to the caller, every call logged on your systems | Model gateway: Every prompt passes through a third party ##### Ownership Agentic Router: You own all the code and the data | Model gateway: A subscription #### Benefits That Drive Impact ##### The developer or agency you work with One key and one SDK instead of a vendor account per model and a connector per tool. ##### Platform teams One bill, cost per call, and ceilings that hold without a meeting. ##### Security and leadership Two auth modes under one policy, every call logged, and the whole stack owned rather than rented. #### Part of the ibl.ai platform Agentic Router is how engineers reach the platform: Agentic OS runs the agents, Agentic Ontology exposes your systems over MCP, Agentic Vibe builds the apps that call them, and Agentic LMS, SIS, Wallet and CRM are applications built the same way. One key, one bill, one platform you own. #### Experience Agentic Router Today ##### Start for Free Start at ibl.ai/join, open an agent in Agentic OS, and create a platform API key on its API tab ##### Read the skills 161 agent skills documented at ibl.ai/developer/skills, one per capability, each mapped to exact REST calls, for Claude Code, Codex, Cursor and any skills-compatible agent ##### Read the API reference Every endpoint, live from the platform's OpenAPI document, at ibl.ai/developer/api-reference ##### Schedule a Demo Walk through your integration, your models and your cost controls with our engineers ### K-12 > Source: https://ibl.ai/product/agentic-router/k12 One API key for a district's IT team and ed-tech partners: every model, 1,000+ MCP servers and the district's agents, callable from the family portal or classroom tools, per user, with one bill and spend caps. On ibl.ai you own all the code and the data, run it model-agnostic across any LLM, and pay with no per-seat pricing — so you can deploy anywhere, from your own cloud to a fully air-gapped network. Agentic Router by ibl.ai gives a district's developers one key to everything the platform serves: 45 providers and 316 chat models, the MCP tools that reach PowerSchool and Google Classroom in place, and the agents the district has already built. Call it from the family portal, the LMS or a partner's tool, with each teacher's or family member's own token, under the district's data governance and spend caps. Deployed in your infrastructure with your keys and your controls—including full code access. #### For The district's IT team and ed-tech partners: One Key, Every Model and Tool ##### Every model, one endpoint 45 providers and 316 chat models on one grid as of November 6, 2026: Anthropic, Google, OpenAI, Meta, Mistral, DeepSeek, Qwen, xAI and the open-weight labs, through OpenRouter, Vertex, Foundry, Bedrock and each provider's own API. Switch a model per request or per agent; prompts, data and connectors stay. ##### Every tool, over MCP Register any of the 1,000+ public MCP servers or your own, attach it to an agent, and bind credentials at the scope you choose: the whole organization, one agent or one user. The tools reach PowerSchool, Google Classroom and ParentSquare in place, with no copy of the data. ##### Your own agents, from your own code Every agent your organization has built on Agentic OS is callable from the family portal, the LMS and classroom tools: a REST call, a streaming WebSocket session, the SDK, an agent skill or any MCP client. For example: "Add the family-communication agent to the parent portal, with each family's own token." ##### Result One integration instead of a vendor account per model and a connector per tool. #### For Platform Teams: One Bill, Enforced Ceilings ##### Unified billing Every model and every tool behind the key lands on one bill, priced by usage, never by seat. The Cost tab breaks spend down by provider, model, user and individual call, with latency percentiles beside it. ##### Spend caps at three scopes A ceiling for the whole organization, for one agent, and for one person on one agent, each with its own reset period and a mode that blocks or alerts at the limit. Enforcement is server-side, so a runaway script stops at the cap. ##### Your keys first A request routes to the highest-priority gateway that can serve the model and has a usable key, and your organization's own provider keys are preferred over platform keys. A deployment with provider keys and no ibl.ai key works with no extra configuration. ##### A catalogue that keeps itself current The model list syncs every six hours and a new model goes live only after a verification gate has tested multi-turn recall, streaming, tool calling, image input and a real, non-zero cost. For example: "Route lesson-plan drafts to a model the district has approved, and change it district-wide in one place." ##### Result Cost you can see per call and cap per scope, with no procurement cycle per model. #### For Security & IT: Two Ways to Authenticate, One Place to Audit ##### Per user Teachers, staff and families sign in with the identity they already have, and every call carries that person's own token. Roles are enforced on each request, so an agent answers only what the caller may see, and a person's in-chat model choice stays inside the list you allow. ##### Server to server A tenant-scoped platform API key, sent as an Api-Token header, authenticates your backend: mint per-user tokens, provision users, run pipelines. It can act for any user on the tenant, so it lives on your server, never in browser code. ##### Where it runs Your cloud, on-premise or fully air-gapped, where the same grid points at models served inside the perimeter. For example: "Cap what any one classroom tool can spend per month." ##### Every call logged Who asked, which model answered, which tools were called and what it cost, on systems you administer, with replayable transcripts and an audit trail per agent. ##### Result One key policy for people and for services, and nothing to reconcile across vendors. #### How Do You Call Agentic Router? ##### REST More than 1,000 documented endpoints: agents, sessions, datasets, memory, MCP servers and connections, users, roles, analytics and billing. Authorization: Api-Token for a service, or the user's own token. ##### Streaming Chat sessions stream over WebSocket, so a response renders token by token in your own interface. ##### SDK and CLI @iblai/iblai-js for web and native apps, @iblai/cli for the terminal, and a Python quickstart (iblai/quickstarts) that creates an agent and chats with it from a script. ##### Agent skills 161 skills from iblai/vibe teach Claude Code, Codex, Cursor, OpenCode or any skills-compatible coding agent to run the platform itself: configure agents, manage datasets and memory, pull cost analytics, each mapped to exact REST calls. Every skill is documented at ibl.ai/developer/skills; install with npx -y @iblai/mcp. ##### Any MCP client Every agent you deploy can be added to Claude Code, Cursor, Claude Desktop or any other MCP client in one step, and talked to from there with its own knowledge, memory and guardrails. ##### Embed and proxy Embed the chat widget in your own app with tokens provisioned from your backend, no login popup. Third-party services such as ElevenLabs and HeyGen sit behind the same key through the external service proxy. #### What's Inside Agentic Router (Highlights) ##### Model catalogue 45 providers, 316 chat models, synced every six hours behind a verification gate ##### Gateways OpenRouter, Vertex, Foundry, Bedrock and each provider's own API, your keys preferred ##### MCP servers 1,000+ public servers plus your own; credentials bound per organization, agent or user; OAuth-backed connectors ##### Your agents Every agent on Agentic OS callable by ID, with its knowledge, memory and guardrails ##### Two auth modes Per-user tokens with roles enforced, or a tenant-scoped platform API key for services ##### Spend caps Organization, agent and per-user ceilings; block or alert; reset periods ##### Cost per call Spend by provider, model, user and call, with latency percentiles ##### Surfaces REST, WebSocket, @iblai/iblai-js, @iblai/cli, 161 agent skills for Claude Code, Codex and Cursor, any MCP client, embed, external service proxy #### Why Agentic Router Instead of a Model Gateway? ##### What the key reaches Agentic Router: Every model, every MCP tool and every agent you own | Model gateway: Models only ##### Authentication Agentic Router: Per-user tokens with roles, or a tenant-scoped platform key | Model gateway: One organization key for everyone ##### Billing Agentic Router: One bill across models and tools, caps per organization, agent and user | Model gateway: Per-model spend, one cap if any ##### Where it runs Agentic Router: Your cloud, on-premise or air-gapped, your provider keys first | Model gateway: The vendor's cloud, through the vendor's keys ##### Your data Agentic Router: Read in place over MCP, scoped to the caller, every call logged on your systems | Model gateway: Every prompt passes through a third party ##### Ownership Agentic Router: You own all the code and the data | Model gateway: A subscription #### Benefits That Drive Impact ##### The district's IT team and ed-tech partners One key and one SDK instead of a vendor account per model and a connector per tool. ##### Platform teams One bill, cost per call, and ceilings that hold without a meeting. ##### Security and leadership Two auth modes under one policy, every call logged, and the whole stack owned rather than rented. #### Part of the ibl.ai platform Agentic Router is how engineers reach the platform: Agentic OS runs the agents, Agentic Ontology exposes your systems over MCP, Agentic Vibe builds the apps that call them, and Agentic LMS, SIS, Wallet and CRM are applications built the same way. One key, one bill, one platform you own. #### Experience Agentic Router Today ##### Start for Free Start at ibl.ai/join, open an agent in Agentic OS, and create a platform API key on its API tab ##### Read the skills 161 agent skills documented at ibl.ai/developer/skills, one per capability, each mapped to exact REST calls, for Claude Code, Codex, Cursor and any skills-compatible agent ##### Read the API reference Every endpoint, live from the platform's OpenAPI document, at ibl.ai/developer/api-reference ##### Schedule a Demo Walk through your integration, your models and your cost controls with our engineers ### Government > Source: https://ibl.ai/product/agentic-router/government One API key for agency developers and integrators: every model, 1,000+ MCP servers and the agency's agents behind one endpoint, per staff member or server to server, on GovCloud, on-premise or air-gapped, with spend caps. On ibl.ai you own all the code and the data, run it model-agnostic across any LLM, and pay with no per-seat pricing — so you can deploy anywhere, from your own cloud to a fully air-gapped network. Agentic Router by ibl.ai gives an agency's developers one key to everything the platform serves: 45 providers and 316 chat models, including models served inside the perimeter, the MCP tools that reach case management and records in place, and the agents the agency has already built. Call it from a constituent portal, a case system or a batch job, with each staff member's own token or a platform key, every call logged and capped. Deployed on GovCloud, on-premise or air-gapped, with your keys and your controls—including full code access. #### For Agency developers and integrators: One Key, Every Model and Tool ##### Every model, one endpoint 45 providers and 316 chat models on one grid as of November 6, 2026: Anthropic, Google, OpenAI, Meta, Mistral, DeepSeek, Qwen, xAI and the open-weight labs, through OpenRouter, Vertex, Foundry, Bedrock and each provider's own API. Switch a model per request or per agent; prompts, data and connectors stay. ##### Every tool, over MCP Register any of the 1,000+ public MCP servers or your own, attach it to an agent, and bind credentials at the scope you choose: the whole organization, one agent or one user. The tools reach ServiceNow, records management and Microsoft Entra ID in place, with no copy of the data. ##### Your own agents, from your own code Every agent your organization has built on Agentic OS is callable from constituent portals, case systems and internal tools: a REST call, a streaming WebSocket session, the SDK, an agent skill or any MCP client. For example: "Put the citizen-services agent behind the agency's web portal, with each caseworker's own token." ##### Result One integration instead of a vendor account per model and a connector per tool. #### For Platform Teams: One Bill, Enforced Ceilings ##### Unified billing Every model and every tool behind the key lands on one bill, priced by usage, never by seat. The Cost tab breaks spend down by provider, model, user and individual call, with latency percentiles beside it. ##### Spend caps at three scopes A ceiling for the whole organization, for one agent, and for one person on one agent, each with its own reset period and a mode that blocks or alerts at the limit. Enforcement is server-side, so a runaway script stops at the cap. ##### Your keys first A request routes to the highest-priority gateway that can serve the model and has a usable key, and your organization's own provider keys are preferred over platform keys. A deployment with provider keys and no ibl.ai key works with no extra configuration. ##### A catalogue that keeps itself current The model list syncs every six hours and a new model goes live only after a verification gate has tested multi-turn recall, streaming, tool calling, image input and a real, non-zero cost. For example: "Point the same grid at models served inside the perimeter for an air-gapped deployment." ##### Result Cost you can see per call and cap per scope, with no procurement cycle per model. #### For Security & IT: Two Ways to Authenticate, One Place to Audit ##### Per user Staff and constituents sign in with the identity they already have, and every call carries that person's own token. Roles are enforced on each request, so an agent answers only what the caller may see, and a person's in-chat model choice stays inside the list you allow. ##### Server to server A tenant-scoped platform API key, sent as an Api-Token header, authenticates your backend: mint per-user tokens, provision users, run pipelines. It can act for any user on the tenant, so it lives on your server, never in browser code. ##### Where it runs Your cloud, on-premise or fully air-gapped, where the same grid points at models served inside the perimeter. For example: "Give a records pipeline a platform key capped per month, with every call in the audit log." ##### Every call logged Who asked, which model answered, which tools were called and what it cost, on systems you administer, with replayable transcripts and an audit trail per agent. ##### Result One key policy for people and for services, and nothing to reconcile across vendors. #### How Do You Call Agentic Router? ##### REST More than 1,000 documented endpoints: agents, sessions, datasets, memory, MCP servers and connections, users, roles, analytics and billing. Authorization: Api-Token for a service, or the user's own token. ##### Streaming Chat sessions stream over WebSocket, so a response renders token by token in your own interface. ##### SDK and CLI @iblai/iblai-js for web and native apps, @iblai/cli for the terminal, and a Python quickstart (iblai/quickstarts) that creates an agent and chats with it from a script. ##### Agent skills 161 skills from iblai/vibe teach Claude Code, Codex, Cursor, OpenCode or any skills-compatible coding agent to run the platform itself: configure agents, manage datasets and memory, pull cost analytics, each mapped to exact REST calls. Every skill is documented at ibl.ai/developer/skills; install with npx -y @iblai/mcp. ##### Any MCP client Every agent you deploy can be added to Claude Code, Cursor, Claude Desktop or any other MCP client in one step, and talked to from there with its own knowledge, memory and guardrails. ##### Embed and proxy Embed the chat widget in your own app with tokens provisioned from your backend, no login popup. Third-party services such as ElevenLabs and HeyGen sit behind the same key through the external service proxy. #### What's Inside Agentic Router (Highlights) ##### Model catalogue 45 providers, 316 chat models, synced every six hours behind a verification gate ##### Gateways OpenRouter, Vertex, Foundry, Bedrock and each provider's own API, your keys preferred ##### MCP servers 1,000+ public servers plus your own; credentials bound per organization, agent or user; OAuth-backed connectors ##### Your agents Every agent on Agentic OS callable by ID, with its knowledge, memory and guardrails ##### Two auth modes Per-user tokens with roles enforced, or a tenant-scoped platform API key for services ##### Spend caps Organization, agent and per-user ceilings; block or alert; reset periods ##### Cost per call Spend by provider, model, user and call, with latency percentiles ##### Surfaces REST, WebSocket, @iblai/iblai-js, @iblai/cli, 161 agent skills for Claude Code, Codex and Cursor, any MCP client, embed, external service proxy #### Why Agentic Router Instead of a Model Gateway? ##### What the key reaches Agentic Router: Every model, every MCP tool and every agent you own | Model gateway: Models only ##### Authentication Agentic Router: Per-user tokens with roles, or a tenant-scoped platform key | Model gateway: One organization key for everyone ##### Billing Agentic Router: One bill across models and tools, caps per organization, agent and user | Model gateway: Per-model spend, one cap if any ##### Where it runs Agentic Router: Your cloud, on-premise or air-gapped, your provider keys first | Model gateway: The vendor's cloud, through the vendor's keys ##### Your data Agentic Router: Read in place over MCP, scoped to the caller, every call logged on your systems | Model gateway: Every prompt passes through a third party ##### Ownership Agentic Router: You own all the code and the data | Model gateway: A subscription #### Benefits That Drive Impact ##### Agency developers and integrators One key and one SDK instead of a vendor account per model and a connector per tool. ##### Platform teams One bill, cost per call, and ceilings that hold without a meeting. ##### Security and leadership Two auth modes under one policy, every call logged, and the whole stack owned rather than rented. #### Part of the ibl.ai platform Agentic Router is how engineers reach the platform: Agentic OS runs the agents, Agentic Ontology exposes your systems over MCP, Agentic Vibe builds the apps that call them, and Agentic LMS, SIS, Wallet and CRM are applications built the same way. One key, one bill, one platform you own. #### Experience Agentic Router Today ##### Start for Free Start at ibl.ai/join, open an agent in Agentic OS, and create a platform API key on its API tab ##### Read the skills 161 agent skills documented at ibl.ai/developer/skills, one per capability, each mapped to exact REST calls, for Claude Code, Codex, Cursor and any skills-compatible agent ##### Read the API reference Every endpoint, live from the platform's OpenAPI document, at ibl.ai/developer/api-reference ##### Schedule a Demo Walk through your integration, your models and your cost controls with our engineers ### Legal > Source: https://ibl.ai/product/agentic-router/legal One API key for a firm's legal-ops and IT developers: every model, 1,000+ MCP servers and the firm's agents, callable from the matter system or the client portal, per attorney or server to server, with one bill and spend caps. On ibl.ai you own all the code and the data, run it model-agnostic across any LLM, and pay with no per-seat pricing — so you can deploy anywhere, from your own cloud to a fully air-gapped network. Agentic Router by ibl.ai gives a firm's developers one key to everything the platform serves: 45 providers and 316 chat models, the MCP tools that reach Clio, iManage and DocuSign in place, and the agents the firm has already built. Call it from the matter system, intake or the client portal, with each attorney's own token or a platform key, with privilege-aware roles enforced on every call. Deployed in your infrastructure with your keys and your controls—including full code access. #### For The firm's legal-ops and IT developers: One Key, Every Model and Tool ##### Every model, one endpoint 45 providers and 316 chat models on one grid as of November 6, 2026: Anthropic, Google, OpenAI, Meta, Mistral, DeepSeek, Qwen, xAI and the open-weight labs, through OpenRouter, Vertex, Foundry, Bedrock and each provider's own API. Switch a model per request or per agent; prompts, data and connectors stay. ##### Every tool, over MCP Register any of the 1,000+ public MCP servers or your own, attach it to an agent, and bind credentials at the scope you choose: the whole organization, one agent or one user. The tools reach Clio, iManage and DocuSign in place, with no copy of the data. ##### Your own agents, from your own code Every agent your organization has built on Agentic OS is callable from the matter system, intake and the client portal: a REST call, a streaming WebSocket session, the SDK, an agent skill or any MCP client. For example: "Run the conflicts-check agent from the intake form, server to server, before a matter opens." ##### Result One integration instead of a vendor account per model and a connector per tool. #### For Platform Teams: One Bill, Enforced Ceilings ##### Unified billing Every model and every tool behind the key lands on one bill, priced by usage, never by seat. The Cost tab breaks spend down by provider, model, user and individual call, with latency percentiles beside it. ##### Spend caps at three scopes A ceiling for the whole organization, for one agent, and for one person on one agent, each with its own reset period and a mode that blocks or alerts at the limit. Enforcement is server-side, so a runaway script stops at the cap. ##### Your keys first A request routes to the highest-priority gateway that can serve the model and has a usable key, and your organization's own provider keys are preferred over platform keys. A deployment with provider keys and no ibl.ai key works with no extra configuration. ##### A catalogue that keeps itself current The model list syncs every six hours and a new model goes live only after a verification gate has tested multi-turn recall, streaming, tool calling, image input and a real, non-zero cost. For example: "Send first-pass contract review to one model and a client-facing summary to another, from one key." ##### Result Cost you can see per call and cap per scope, with no procurement cycle per model. #### For Security & IT: Two Ways to Authenticate, One Place to Audit ##### Per user Attorneys, staff and clients sign in with the identity they already have, and every call carries that person's own token. Roles are enforced on each request, so an agent answers only what the caller may see, and a person's in-chat model choice stays inside the list you allow. ##### Server to server A tenant-scoped platform API key, sent as an Api-Token header, authenticates your backend: mint per-user tokens, provision users, run pipelines. It can act for any user on the tenant, so it lives on your server, never in browser code. ##### Where it runs Your cloud, on-premise or fully air-gapped, where the same grid points at models served inside the perimeter. For example: "Cap each practice group's monthly spend and see cost per matter." ##### Every call logged Who asked, which model answered, which tools were called and what it cost, on systems you administer, with replayable transcripts and an audit trail per agent. ##### Result One key policy for people and for services, and nothing to reconcile across vendors. #### How Do You Call Agentic Router? ##### REST More than 1,000 documented endpoints: agents, sessions, datasets, memory, MCP servers and connections, users, roles, analytics and billing. Authorization: Api-Token for a service, or the user's own token. ##### Streaming Chat sessions stream over WebSocket, so a response renders token by token in your own interface. ##### SDK and CLI @iblai/iblai-js for web and native apps, @iblai/cli for the terminal, and a Python quickstart (iblai/quickstarts) that creates an agent and chats with it from a script. ##### Agent skills 161 skills from iblai/vibe teach Claude Code, Codex, Cursor, OpenCode or any skills-compatible coding agent to run the platform itself: configure agents, manage datasets and memory, pull cost analytics, each mapped to exact REST calls. Every skill is documented at ibl.ai/developer/skills; install with npx -y @iblai/mcp. ##### Any MCP client Every agent you deploy can be added to Claude Code, Cursor, Claude Desktop or any other MCP client in one step, and talked to from there with its own knowledge, memory and guardrails. ##### Embed and proxy Embed the chat widget in your own app with tokens provisioned from your backend, no login popup. Third-party services such as ElevenLabs and HeyGen sit behind the same key through the external service proxy. #### What's Inside Agentic Router (Highlights) ##### Model catalogue 45 providers, 316 chat models, synced every six hours behind a verification gate ##### Gateways OpenRouter, Vertex, Foundry, Bedrock and each provider's own API, your keys preferred ##### MCP servers 1,000+ public servers plus your own; credentials bound per organization, agent or user; OAuth-backed connectors ##### Your agents Every agent on Agentic OS callable by ID, with its knowledge, memory and guardrails ##### Two auth modes Per-user tokens with roles enforced, or a tenant-scoped platform API key for services ##### Spend caps Organization, agent and per-user ceilings; block or alert; reset periods ##### Cost per call Spend by provider, model, user and call, with latency percentiles ##### Surfaces REST, WebSocket, @iblai/iblai-js, @iblai/cli, 161 agent skills for Claude Code, Codex and Cursor, any MCP client, embed, external service proxy #### Why Agentic Router Instead of a Model Gateway? ##### What the key reaches Agentic Router: Every model, every MCP tool and every agent you own | Model gateway: Models only ##### Authentication Agentic Router: Per-user tokens with roles, or a tenant-scoped platform key | Model gateway: One organization key for everyone ##### Billing Agentic Router: One bill across models and tools, caps per organization, agent and user | Model gateway: Per-model spend, one cap if any ##### Where it runs Agentic Router: Your cloud, on-premise or air-gapped, your provider keys first | Model gateway: The vendor's cloud, through the vendor's keys ##### Your data Agentic Router: Read in place over MCP, scoped to the caller, every call logged on your systems | Model gateway: Every prompt passes through a third party ##### Ownership Agentic Router: You own all the code and the data | Model gateway: A subscription #### Benefits That Drive Impact ##### The firm's legal-ops and IT developers One key and one SDK instead of a vendor account per model and a connector per tool. ##### Platform teams One bill, cost per call, and ceilings that hold without a meeting. ##### Security and leadership Two auth modes under one policy, every call logged, and the whole stack owned rather than rented. #### Part of the ibl.ai platform Agentic Router is how engineers reach the platform: Agentic OS runs the agents, Agentic Ontology exposes your systems over MCP, Agentic Vibe builds the apps that call them, and Agentic LMS, SIS, Wallet and CRM are applications built the same way. One key, one bill, one platform you own. #### Experience Agentic Router Today ##### Start for Free Start at ibl.ai/join, open an agent in Agentic OS, and create a platform API key on its API tab ##### Read the skills 161 agent skills documented at ibl.ai/developer/skills, one per capability, each mapped to exact REST calls, for Claude Code, Codex, Cursor and any skills-compatible agent ##### Read the API reference Every endpoint, live from the platform's OpenAPI document, at ibl.ai/developer/api-reference ##### Schedule a Demo Walk through your integration, your models and your cost controls with our engineers ### Financial Services > Source: https://ibl.ai/product/agentic-router/financial-services One API key for a bank's platform engineering team: every model, 1,000+ MCP servers and the institution's agents behind one endpoint, per advisor or server to server, with unified billing, spend caps and an audit trail per call. On ibl.ai you own all the code and the data, run it model-agnostic across any LLM, and pay with no per-seat pricing — so you can deploy anywhere, from your own cloud to a fully air-gapped network. Agentic Router by ibl.ai gives platform engineering one key to everything the platform serves: 45 providers and 316 chat models, the MCP tools that reach core banking, Salesforce and Actimize in place, and the agents the institution has already built. Call it from the advisor desktop, onboarding or a servicing app, with each advisor's own token or a platform key, every call logged with its cost. Deployed in your infrastructure with your keys and your controls—including full code access. #### For Platform engineering and integration teams: One Key, Every Model and Tool ##### Every model, one endpoint 45 providers and 316 chat models on one grid as of November 6, 2026: Anthropic, Google, OpenAI, Meta, Mistral, DeepSeek, Qwen, xAI and the open-weight labs, through OpenRouter, Vertex, Foundry, Bedrock and each provider's own API. Switch a model per request or per agent; prompts, data and connectors stay. ##### Every tool, over MCP Register any of the 1,000+ public MCP servers or your own, attach it to an agent, and bind credentials at the scope you choose: the whole organization, one agent or one user. The tools reach core banking, Salesforce Financial Services Cloud and NICE Actimize in place, with no copy of the data. ##### Your own agents, from your own code Every agent your organization has built on Agentic OS is callable from the advisor desktop, onboarding and servicing apps: a REST call, a streaming WebSocket session, the SDK, an agent skill or any MCP client. For example: "Run the KYC agent from the onboarding flow, server to server, with the file assembled before a person reviews it." ##### Result One integration instead of a vendor account per model and a connector per tool. #### For Platform Teams: One Bill, Enforced Ceilings ##### Unified billing Every model and every tool behind the key lands on one bill, priced by usage, never by seat. The Cost tab breaks spend down by provider, model, user and individual call, with latency percentiles beside it. ##### Spend caps at three scopes A ceiling for the whole organization, for one agent, and for one person on one agent, each with its own reset period and a mode that blocks or alerts at the limit. Enforcement is server-side, so a runaway script stops at the cap. ##### Your keys first A request routes to the highest-priority gateway that can serve the model and has a usable key, and your organization's own provider keys are preferred over platform keys. A deployment with provider keys and no ibl.ai key works with no extra configuration. ##### A catalogue that keeps itself current The model list syncs every six hours and a new model goes live only after a verification gate has tested multi-turn recall, streaming, tool calling, image input and a real, non-zero cost. For example: "Keep client-facing drafting on an approved model and let research use a cheaper one." ##### Result Cost you can see per call and cap per scope, with no procurement cycle per model. #### For Security & IT: Two Ways to Authenticate, One Place to Audit ##### Per user Advisors, operations staff and clients sign in with the identity they already have, and every call carries that person's own token. Roles are enforced on each request, so an agent answers only what the caller may see, and a person's in-chat model choice stays inside the list you allow. ##### Server to server A tenant-scoped platform API key, sent as an Api-Token header, authenticates your backend: mint per-user tokens, provision users, run pipelines. It can act for any user on the tenant, so it lives on your server, never in browser code. ##### Where it runs Your cloud, on-premise or fully air-gapped, where the same grid points at models served inside the perimeter. For example: "Cap spend per desk and reconcile cost per call against the ledger." ##### Every call logged Who asked, which model answered, which tools were called and what it cost, on systems you administer, with replayable transcripts and an audit trail per agent. ##### Result One key policy for people and for services, and nothing to reconcile across vendors. #### How Do You Call Agentic Router? ##### REST More than 1,000 documented endpoints: agents, sessions, datasets, memory, MCP servers and connections, users, roles, analytics and billing. Authorization: Api-Token for a service, or the user's own token. ##### Streaming Chat sessions stream over WebSocket, so a response renders token by token in your own interface. ##### SDK and CLI @iblai/iblai-js for web and native apps, @iblai/cli for the terminal, and a Python quickstart (iblai/quickstarts) that creates an agent and chats with it from a script. ##### Agent skills 161 skills from iblai/vibe teach Claude Code, Codex, Cursor, OpenCode or any skills-compatible coding agent to run the platform itself: configure agents, manage datasets and memory, pull cost analytics, each mapped to exact REST calls. Every skill is documented at ibl.ai/developer/skills; install with npx -y @iblai/mcp. ##### Any MCP client Every agent you deploy can be added to Claude Code, Cursor, Claude Desktop or any other MCP client in one step, and talked to from there with its own knowledge, memory and guardrails. ##### Embed and proxy Embed the chat widget in your own app with tokens provisioned from your backend, no login popup. Third-party services such as ElevenLabs and HeyGen sit behind the same key through the external service proxy. #### What's Inside Agentic Router (Highlights) ##### Model catalogue 45 providers, 316 chat models, synced every six hours behind a verification gate ##### Gateways OpenRouter, Vertex, Foundry, Bedrock and each provider's own API, your keys preferred ##### MCP servers 1,000+ public servers plus your own; credentials bound per organization, agent or user; OAuth-backed connectors ##### Your agents Every agent on Agentic OS callable by ID, with its knowledge, memory and guardrails ##### Two auth modes Per-user tokens with roles enforced, or a tenant-scoped platform API key for services ##### Spend caps Organization, agent and per-user ceilings; block or alert; reset periods ##### Cost per call Spend by provider, model, user and call, with latency percentiles ##### Surfaces REST, WebSocket, @iblai/iblai-js, @iblai/cli, 161 agent skills for Claude Code, Codex and Cursor, any MCP client, embed, external service proxy #### Why Agentic Router Instead of a Model Gateway? ##### What the key reaches Agentic Router: Every model, every MCP tool and every agent you own | Model gateway: Models only ##### Authentication Agentic Router: Per-user tokens with roles, or a tenant-scoped platform key | Model gateway: One organization key for everyone ##### Billing Agentic Router: One bill across models and tools, caps per organization, agent and user | Model gateway: Per-model spend, one cap if any ##### Where it runs Agentic Router: Your cloud, on-premise or air-gapped, your provider keys first | Model gateway: The vendor's cloud, through the vendor's keys ##### Your data Agentic Router: Read in place over MCP, scoped to the caller, every call logged on your systems | Model gateway: Every prompt passes through a third party ##### Ownership Agentic Router: You own all the code and the data | Model gateway: A subscription #### Benefits That Drive Impact ##### Platform engineering and integration teams One key and one SDK instead of a vendor account per model and a connector per tool. ##### Platform teams One bill, cost per call, and ceilings that hold without a meeting. ##### Security and leadership Two auth modes under one policy, every call logged, and the whole stack owned rather than rented. #### Part of the ibl.ai platform Agentic Router is how engineers reach the platform: Agentic OS runs the agents, Agentic Ontology exposes your systems over MCP, Agentic Vibe builds the apps that call them, and Agentic LMS, SIS, Wallet and CRM are applications built the same way. One key, one bill, one platform you own. #### Experience Agentic Router Today ##### Start for Free Start at ibl.ai/join, open an agent in Agentic OS, and create a platform API key on its API tab ##### Read the skills 161 agent skills documented at ibl.ai/developer/skills, one per capability, each mapped to exact REST calls, for Claude Code, Codex, Cursor and any skills-compatible agent ##### Read the API reference Every endpoint, live from the platform's OpenAPI document, at ibl.ai/developer/api-reference ##### Schedule a Demo Walk through your integration, your models and your cost controls with our engineers ### Healthcare > Source: https://ibl.ai/product/agentic-router/healthcare One API key for a health system's informatics team: every model, 1,000+ MCP servers and the organization's agents, callable from the EHR sidebar or the patient portal, per clinician or server to server, with PHI never leaving your environment. On ibl.ai you own all the code and the data, run it model-agnostic across any LLM, and pay with no per-seat pricing — so you can deploy anywhere, from your own cloud to a fully air-gapped network. Agentic Router by ibl.ai gives an informatics team one key to everything the platform serves: 45 providers and 316 chat models, the MCP tools that reach the EHR over FHIR and the payer portals in place, and the agents the organization has already built. Call it from the EHR sidebar, the patient portal or revenue-cycle tools, with each clinician's own token or a platform key, with PHI filtered before a model sees it. Deployed in your infrastructure with your keys and your controls—including full code access. #### For Informatics and integration teams: One Key, Every Model and Tool ##### Every model, one endpoint 45 providers and 316 chat models on one grid as of November 6, 2026: Anthropic, Google, OpenAI, Meta, Mistral, DeepSeek, Qwen, xAI and the open-weight labs, through OpenRouter, Vertex, Foundry, Bedrock and each provider's own API. Switch a model per request or per agent; prompts, data and connectors stay. ##### Every tool, over MCP Register any of the 1,000+ public MCP servers or your own, attach it to an agent, and bind credentials at the scope you choose: the whole organization, one agent or one user. The tools reach Epic or Cerner over FHIR, and Availity in place, with no copy of the data. ##### Your own agents, from your own code Every agent your organization has built on Agentic OS is callable from the EHR sidebar, the patient portal and revenue-cycle tools: a REST call, a streaming WebSocket session, the SDK, an agent skill or any MCP client. For example: "Run the prior-authorization agent from the EHR, with each clinician's own token and role." ##### Result One integration instead of a vendor account per model and a connector per tool. #### For Platform Teams: One Bill, Enforced Ceilings ##### Unified billing Every model and every tool behind the key lands on one bill, priced by usage, never by seat. The Cost tab breaks spend down by provider, model, user and individual call, with latency percentiles beside it. ##### Spend caps at three scopes A ceiling for the whole organization, for one agent, and for one person on one agent, each with its own reset period and a mode that blocks or alerts at the limit. Enforcement is server-side, so a runaway script stops at the cap. ##### Your keys first A request routes to the highest-priority gateway that can serve the model and has a usable key, and your organization's own provider keys are preferred over platform keys. A deployment with provider keys and no ibl.ai key works with no extra configuration. ##### A catalogue that keeps itself current The model list syncs every six hours and a new model goes live only after a verification gate has tested multi-turn recall, streaming, tool calling, image input and a real, non-zero cost. For example: "Route coding queries to a model served inside your environment." ##### Result Cost you can see per call and cap per scope, with no procurement cycle per model. #### For Security & IT: Two Ways to Authenticate, One Place to Audit ##### Per user Clinicians, staff and patients sign in with the identity they already have, and every call carries that person's own token. Roles are enforced on each request, so an agent answers only what the caller may see, and a person's in-chat model choice stays inside the list you allow. ##### Server to server A tenant-scoped platform API key, sent as an Api-Token header, authenticates your backend: mint per-user tokens, provision users, run pipelines. It can act for any user on the tenant, so it lives on your server, never in browser code. ##### Where it runs Your cloud, on-premise or fully air-gapped, where the same grid points at models served inside the perimeter. For example: "Cap spend per department and see cost per call for the compliance file." ##### Every call logged Who asked, which model answered, which tools were called and what it cost, on systems you administer, with replayable transcripts and an audit trail per agent. ##### Result One key policy for people and for services, and nothing to reconcile across vendors. #### How Do You Call Agentic Router? ##### REST More than 1,000 documented endpoints: agents, sessions, datasets, memory, MCP servers and connections, users, roles, analytics and billing. Authorization: Api-Token for a service, or the user's own token. ##### Streaming Chat sessions stream over WebSocket, so a response renders token by token in your own interface. ##### SDK and CLI @iblai/iblai-js for web and native apps, @iblai/cli for the terminal, and a Python quickstart (iblai/quickstarts) that creates an agent and chats with it from a script. ##### Agent skills 161 skills from iblai/vibe teach Claude Code, Codex, Cursor, OpenCode or any skills-compatible coding agent to run the platform itself: configure agents, manage datasets and memory, pull cost analytics, each mapped to exact REST calls. Every skill is documented at ibl.ai/developer/skills; install with npx -y @iblai/mcp. ##### Any MCP client Every agent you deploy can be added to Claude Code, Cursor, Claude Desktop or any other MCP client in one step, and talked to from there with its own knowledge, memory and guardrails. ##### Embed and proxy Embed the chat widget in your own app with tokens provisioned from your backend, no login popup. Third-party services such as ElevenLabs and HeyGen sit behind the same key through the external service proxy. #### What's Inside Agentic Router (Highlights) ##### Model catalogue 45 providers, 316 chat models, synced every six hours behind a verification gate ##### Gateways OpenRouter, Vertex, Foundry, Bedrock and each provider's own API, your keys preferred ##### MCP servers 1,000+ public servers plus your own; credentials bound per organization, agent or user; OAuth-backed connectors ##### Your agents Every agent on Agentic OS callable by ID, with its knowledge, memory and guardrails ##### Two auth modes Per-user tokens with roles enforced, or a tenant-scoped platform API key for services ##### Spend caps Organization, agent and per-user ceilings; block or alert; reset periods ##### Cost per call Spend by provider, model, user and call, with latency percentiles ##### Surfaces REST, WebSocket, @iblai/iblai-js, @iblai/cli, 161 agent skills for Claude Code, Codex and Cursor, any MCP client, embed, external service proxy #### Why Agentic Router Instead of a Model Gateway? ##### What the key reaches Agentic Router: Every model, every MCP tool and every agent you own | Model gateway: Models only ##### Authentication Agentic Router: Per-user tokens with roles, or a tenant-scoped platform key | Model gateway: One organization key for everyone ##### Billing Agentic Router: One bill across models and tools, caps per organization, agent and user | Model gateway: Per-model spend, one cap if any ##### Where it runs Agentic Router: Your cloud, on-premise or air-gapped, your provider keys first | Model gateway: The vendor's cloud, through the vendor's keys ##### Your data Agentic Router: Read in place over MCP, scoped to the caller, every call logged on your systems | Model gateway: Every prompt passes through a third party ##### Ownership Agentic Router: You own all the code and the data | Model gateway: A subscription #### Benefits That Drive Impact ##### Informatics and integration teams One key and one SDK instead of a vendor account per model and a connector per tool. ##### Platform teams One bill, cost per call, and ceilings that hold without a meeting. ##### Security and leadership Two auth modes under one policy, every call logged, and the whole stack owned rather than rented. #### Part of the ibl.ai platform Agentic Router is how engineers reach the platform: Agentic OS runs the agents, Agentic Ontology exposes your systems over MCP, Agentic Vibe builds the apps that call them, and Agentic LMS, SIS, Wallet and CRM are applications built the same way. One key, one bill, one platform you own. #### Experience Agentic Router Today ##### Start for Free Start at ibl.ai/join, open an agent in Agentic OS, and create a platform API key on its API tab ##### Read the skills 161 agent skills documented at ibl.ai/developer/skills, one per capability, each mapped to exact REST calls, for Claude Code, Codex, Cursor and any skills-compatible agent ##### Read the API reference Every endpoint, live from the platform's OpenAPI document, at ibl.ai/developer/api-reference ##### Schedule a Demo Walk through your integration, your models and your cost controls with our engineers ## Run Agentic Router on your own terms *Bring us a hard problem. Keep the code.* Start free on our cloud today, or take the whole codebase into your own infrastructure. Either way our engineers work alongside your team, and you own all the code and the data. [Get Started Free](https://ibl.ai/join) · [Book a 30-minute demo](https://cal.com/iblai/30min) · [Read the skills](https://ibl.ai/developer/skills) ibl.ai is family-owned and operated from New York, NY — a U.S.-headquartered, domestically-owned long-term partner, not a vendor that sells licenses and moves on. · SOC 2 Type II certified --- *[View on ibl.ai](https://ibl.ai/product/agentic-router)*