Agentic Index

AI agent platforms that let you bring your own model or route between providers (2026)

The Agentic Index grades 554 agentic AI platforms against the same 14 capabilities as every other lane. These are independent ratings of AI agent platforms, rated from each vendor's own public evidence rather than from analyst opinion, vendor briefings or paid placement. 213 of the 554 agentic AI platforms in the Agentic Index document bringing your own model, bringing your own key or routing tasks between providers in full. That is 38.4% of the platform field. Another 73 document some model choice without enough evidence to clear, 260 document none at all, and 8 could not be verified either way, which accounts for all 554. This is a bar, not a leaderboard: a platform either documents model choice as a buyer facing surface in its own public materials or it does not clear, and Partial evidence does not count. Every platform that clears is named below, ordered by total coverage across all 14 capabilities.

The finding underneath the list is the part a buyer cannot get anywhere else. Model independence and enterprise governance travel together, and even so fewer than one platform in six documents both. Only 82 of the 554 platforms in the index, 14.8%, document both model choice and the four capability enterprise governance bar used on the enterprise security and compliance page. A team that needs both is shopping that field, not this one.

It would be easy to read that as a trade off, and it is not one. The association runs the other way. Documenting model choice makes a platform about twice as likely to clear enterprise governance, not less likely. Across the whole pool, 118 platforms clear governance, a base rate of 21.3%. Among the 213 platforms on this page the rate is 38.5%. It holds in reverse too: 69.5% of the 118 governance cleared platforms document model choice, against a field base rate of 38.4%. If the two capabilities were independent the overlap would be about 45 platforms. It is 82. The vendors that expose the model are, on documented evidence, the same vendors that expose the audit log and the deployment target.

So the scarcity sits in the joint requirement rather than in any tension between the two, and it is real scarcity. 131 of the 213 platforms on this page still do not clear governance, and 36 of the 118 governance cleared platforms still do not document model choice. Doubling a small number leaves a small number. A buyer who has already shortlisted on governance has ruled out most of this list without meaning to, and a buyer who shortlisted on model independence has done the same in reverse, which is why the two requirements are worth stating together at the start of an evaluation rather than discovering in sequence.

The same shape shows up on deployment, which is the narrower question most buyers actually mean. 171 of these 213 platforms also run inside your own boundary, well above the pool base rate, and 42 will take your model while the software keeps running in the vendor's cloud. Which of those two shapes is the right answer depends on whether your constraint is which model sees the data or where the software runs, and the two are separate purchases. The self hosting page counts the same intersection from the deployment side, where the gap runs the other way.

The bar, and what the 213 platforms that clear it also document

Model flexibility and routing at Full is the whole bar on this page: the platform documents bringing your own model or bringing your own key (BYOK), or routing tasks between models or providers, rather than being locked to one vendor hosted model. The table below is not the bar. It shows how the 213 platforms that clear it score on enterprise governance, next to the rate across the whole pool.

Governance capability	What has to be documented	Pool	Also on this list
Security and identity governance	SOC 2 or ISO 27001, single sign on (SSO or SAML) and role based access control (RBAC)	335 (60%)	161 (76%)
Deployment and data residency	self hosted, on premises, air gapped or private VPC deployment, or documented control over where agent data sits	281 (51%)	171 (80%)
Observability and auditability	tracing and audit logging across agent runs	326 (59%)	153 (72%)
Human oversight and guardrails	a human approval step before an agent action executes	292 (53%)	120 (56%)

Full means the vendor publishes evidence meeting the capability in its own public materials, under the Agentic Index verification standard. Pool counts the 554 platforms in the index. Also on this list counts the 213 that clear the model bar and document that capability too. Every row runs higher in the right hand column than in the left, which is the positive association stated above holding capability by capability rather than only on the combined bar. One honest limit on the bar itself: this axis grades bring your own key and routing between providers as a single capability, so a Full grade does not by itself separate the two. Read the entry and the vendor profile behind it before assuming which one a platform documents.

Clears the bar and scores 12.5 or higher of 14 overall

These 40 platforms document model flexibility and routing in full and also sit at the top of the Agentic Index coverage score across all 14 capabilities. Ordered by total coverage, ties broken alphabetically.

1. 1.Appian

   14.0 / 14 capabilities

   enterprise operations agents, agent builders

   Model choice is set per agent in its model selection, with AUTO as the default taking the latest model from the configured provider, so a builder can pin a model on one agent while others follow the provider's newest.

2. 2.FLOWX.AI

   14.0 / 14 capabilities

   enterprise operations agents, multi-agent platforms, agent builders, agent infrastructure platforms

   Enterprise agentic platform whose Model Providers settings let the customer add a provider from a catalog, supply their own API key or use the FLOWX managed key, and test the connection before use. The vendor answers who chooses the model in its own documentation, which is the whole bar.

3. 3.Gumloop

   14.0 / 14 capabilities

   agent builders

   Visual no code canvas where model choice is a buyer facing surface: Agent Preferences sits at the top of each agent's configuration with Recommended presets and the full provider list, and new models are usually available within a day of public release. The one stated constraint is tool calling, which keeps models without it in workflow nodes rather than in the agent picker.

4. 4.Mastra

   14.0 / 14 capabilities

   agent infrastructure platforms

   Each agent's model is a provider and model string in Mastra's model router, which reaches thousands of models across OpenAI, Anthropic, Google and others, and the Platform gateway accepts the customer's own keys.

5. 5.Microsoft

   14.0 / 14 capabilities

   enterprise operations agents

   The maker chooses the primary model for generative orchestration from a dropdown, with separate settings for deep reasoning, generative responses and the prompt builder, an external model as a primary option, and published availability tables by region, with experimental models tagged.

6. 6.ServiceNow

   14.0 / 14 capabilities

   enterprise operations agents, multi-agent platforms

   An administrator selects the default provider for AI agents in AI Admin Hub from Now LLM Service, AWS Claude, Azure OpenAI and Google Gemini, testing models in the skill kit first, and ServiceNow advises retesting agents after a global switch.

7. 7.UiPath

   14.0 / 14 capabilities

   enterprise operations agents, multi-agent platforms, agent builders

   Agentic automation orchestrating agents, robots and people end to end. Autopilot supports GPT, Gemini, Claude and any compatible model, run on the customer's cloud of choice or their own infrastructure, with an LLM selector in Agent Builder. The highest verified coverage score in the index and it does not tie you to a model.

8. 8.Pega

   13.5 / 14 capabilities

   enterprise operations agents, agent builders

   Provider agnostic at the model layer across Azure OpenAI, AWS Bedrock and Google Vertex, with the provider named as a design decision for the customer's lead architect, weighing cost, performance and compliance.

9. 9.StackAI

   13.5 / 14 capabilities

   enterprise operations agents, agent builders

   No code enterprise platform documenting model choice as a governed surface rather than a dropdown: an LLM hosting tree covering multiple providers, models running in the customer's own Azure or AWS Bedrock account, and Local LLM for models on the customer's own hardware, with an LLM provider governance page controlling which models the organization can reach and where the data goes.

10. 10.Agno

    13.0 / 14 capabilities

    agent infrastructure platforms

    High performance agent runtime, formerly Phidata, Apache 2.0 licensed, which is the strongest structural position for model independence because the runtime is yours to point anywhere. The evidence caveat beside this entry applies to the model row specifically.

11. 11.AutoGPT

    13.0 / 14 capabilities

    multi-agent platforms, agent builders, agent infrastructure platforms

    Model choice is made per block, with OpenAI, Anthropic, Groq and Llama models built in and other providers connected through AI blocks. Self hosting is free with the customer's own model keys, so the model bill runs through the customer's own provider accounts.

12. 12.Base44

    13.0 / 14 capabilities

    agent builders

    Each agent has an AI Model setting: Automatic for fast, low cost responses, or a named model from Google Gemini, OpenAI GPT or Anthropic Claude for harder reasoning, switchable at any time. The choice is per agent rather than per step.

13. 13.CrewAI

    13.0 / 14 capabilities

    multi-agent platforms

    Open source multi agent orchestration for collaborative crews, MIT licensed and Python native, so the model choice is yours in the same sense the code is. Read the evidence caveat beside this entry before treating the grade as thick.

14. 14.Databricks Agent Bricks

    13.0 / 14 capabilities

    multi-agent platforms

    Model choice is governed rather than a dropdown: administrators connect OpenAI, Azure OpenAI, Anthropic, Bedrock, Foundry, Gemini or a custom provider on their own keys as Unity Catalog securables granted per user, alongside hosted open and frontier models.

15. 15.Dataiku

    13.0 / 14 capabilities

    multi-agent platforms, agent builders

    The LLM Mesh is the governed surface: administrators define connections to OpenAI, Azure OpenAI, Anthropic, Bedrock, Vertex, Mistral, Cohere, NIM and others, and locally running Hugging Face models keep inference in house.

16. 16.HappyRobot

    13.0 / 14 capabilities

    enterprise operations agents

    AI workers for operations teams that started in freight, whose model choice is stated on its own pages: workflows can be configured to use OpenAI, Google Gemini, Anthropic Claude or the customer's own hosted models, with HappyRobot handling orchestration and routing between them.

17. 17.n8n

    13.0 / 14 capabilities

    agent builders

    Open source workflow automation where the model is a sub node the customer attaches: the AI Agent node errors without a connected chat model, and because the connection is per node, different agents in one workflow can run different providers. First class nodes cover OpenAI, Anthropic, Gemini, Mistral and Perplexity among others.

18. 18.Relevance AI

    13.0 / 14 capabilities

    multi-agent platforms, agent builders

    Each agent picks its model from several providers, including Claude, OpenAI and Gemini, and can set a backup model from a different provider that retries a failed task; bring your own model keys is available on every plan.

19. 19.Replit Agent

    13.0 / 14 capabilities

    agent builders

    App builder and runtime whose default is intelligent routing, with a documented manual selector and an Effort setting for customers who want the control, and a separate model switch behind the design surface. Read this Full as routing with an override rather than an open provider matrix: manual selection depends on the customer's mode and plan.

20. 20.Sema4.ai

    13.0 / 14 capabilities

    agent builders

    Enterprise agent platform on a stated bring your own model approach naming OpenAI, Azure, Bedrock and Snowflake Cortex, with agents running inside the customer's own cloud account. Snowflake Cortex is the tell: inference sits next to the governed data rather than in the vendor's tenancy.

21. 21.Sim

    13.0 / 14 capabilities

    multi-agent platforms, agent builders

    Open source Apache 2.0 agent workspace and the sharpest single instance on this page: fifteen plus providers documented, bring your own key at base provider pricing with no markup, local models through Ollama or VLLM at no API cost, and enterprise permission groups that restrict which providers are allowed.

22. 22.Tray.ai

    13.0 / 14 capabilities

    agent infrastructure platforms

    Agents run on Tray's native default model or on the customer's own keys for OpenAI or AWS Bedrock, with the provider and the specific model version chosen in the agent dashboard.

23. 23.Tungsten Automation

    13.0 / 14 capabilities

    enterprise operations agents, agent builders

    Bring Your Own Model lets TotalAgility customers combine providers, Azure OpenAI Service among them, with their own custom LLMs.

24. 24.Warp

    13.0 / 14 capabilities

    multi-agent platforms (secondary lane membership, primary category Coding agent)

    A model picker offers Claude, GPT, Gemini, GLM, Kimi and others with an Auto option, tasks can run on different models in parallel, and the model can change mid task. Business adds bring your own API keys and custom inference endpoints.

25. 25.Writer

    13.0 / 14 capabilities

    enterprise operations agents, agent infrastructure platforms

    Enterprise platform built on its own Palmyra model family, and it clears the bar because the vendor answers the model choice question directly in its own FAQ rather than leaving a buyer to infer it. Palmyra is also distributed through Amazon Bedrock, so the models are reachable outside the Writer application.

26. 26.Akka

    12.5 / 14 capabilities

    multi-agent platforms, agent builders, agent infrastructure platforms

    Nine providers are built in, including Anthropic, OpenAI, Gemini, Bedrock, Mistral and Ollama, any other plugs in as a custom provider, and each agent sets its own model.

27. 27.Boomi

    12.5 / 14 capabilities

    enterprise operations agents, multi-agent platforms, agent builders

    Integration platform with an agent control tower over enterprise systems. Model agnostic with MCP support across Amazon Bedrock, Snowflake Cortex and OpenAI, with roadmap support for custom large and small models, and the integration layer is where its evidence comes from.

28. 28.Creatio

    12.5 / 14 capabilities

    enterprise operations agents, agent builders (secondary lane membership, primary category GTM / revenue agent)

    The Prompt Agent Designer lets the builder choose among several LLM providers, with Anthropic Claude, OpenAI and Google Gemini named. Temperature and other model settings are set per agent, inside a registry of approved models. Creatio does not say whether every provider is available on every plan.

29. 29.Fabrix.ai

    12.5 / 14 capabilities

    enterprise operations agents, agent builders

    Multi LLM choice is the first tab of the platform: OpenAI, Anthropic or other public APIs in the cloud, or open source models such as Llama and Mistral deployed inside the customer's infrastructure, with model fallback for vendor neutral orchestration.

30. 30.Harvey

    12.5 / 14 capabilities

    enterprise operations agents

    A legal platform rather than a horizontal one. Workspace admins roll out Anthropic and Google models beside OpenAI through self serve permissions, and an opt in model selector lets users switch models in Assistant, Vault and Workflow Builder while Harvey auto routes by default. Admin managed access plus a user selector is the Full marker.

31. 31.IBM watsonx Orchestrate

    12.5 / 14 capabilities

    enterprise operations agents

    Models from multiple providers are imported as virtual models and assigned per agent, and model policies coordinate several of them with load balance, fallback or single strategies, weights per target and retries on named status codes.

32. 32.Kestra

    12.5 / 14 capabilities

    multi-agent platforms, agent infrastructure platforms

    The customer sets the model on each AI task through a provider property, with Anthropic, OpenAI, Azure OpenAI, Amazon Bedrock, Gemini and Vertex AI, Mistral, DeepSeek, Ollama and OpenRouter among the providers, and an Enterprise policy can fix the provider across a namespace.

33. 33.Latenode

    12.5 / 14 capabilities

    agent builders, agent infrastructure platforms

    Low code automation platform bundling GPT, Claude, Gemini, Deepseek and LLaMA under one subscription, so a customer picks the model per step without holding a provider account for each. The differentiator here is commercial rather than technical: the same choice, without the procurement.

34. 34.Mendix

    12.5 / 14 capabilities

    enterprise operations agents, agent builders

    The customer picks the model for each agent version from a deployed model on any GenAI Commons connector, covering Mendix Cloud GenAI, OpenAI, Amazon Bedrock, Google Gemini, Mistral and Snowflake Cortex. Temperature, top P and max tokens are configurable, and teams can build their own connector.

35. 35.OutSystems

    12.5 / 14 capabilities

    multi-agent platforms, agent builders

    Enterprise low code incumbent whose Agent Workbench connects a model once for reuse across agents, at general availability across Bedrock, Azure OpenAI and Anthropic, plus Gemini, Cohere, Mistral, Databricks, AI2 and IBM watsonx, with custom models on Vertex AI and open source models on Hugging Face.

36. 36.Pipefy

    12.5 / 14 capabilities

    enterprise operations agents, agent builders

    An administrator sets the active LLM provider for the whole organization or one assistant, from Pipefy managed providers or a custom provider the organization registered, and lists the models each provider offers.

37. 37.Salesforce

    12.5 / 14 capabilities

    enterprise operations agents

    CRM platform whose Agentforce layer runs agents across sales, service and marketing. The Atlas Reasoning Engine supports OpenAI, Anthropic and Google Gemini, so an enterprise can pick the model for a governed deployment rather than inherit one. It is also one of the platforms that clears the full governance bar.

38. 38.SimplAI

    12.5 / 14 capabilities

    multi-agent platforms, agent builders, agent infrastructure platforms

    Datafuse Technology Inc builds SimplAI, an enterprise agentic AI operating system spanning cloud, customer cloud, VPC, on premises and air gapped with no architectural change between them. A model layer orchestrates across LLMs with routing and evaluation and the model is selected per agent. One caveat a buyer should carry: bring your own models and fine tuning are documented but gated to the enterprise tier.

39. 39.Torq

    12.5 / 14 capabilities

    enterprise operations agents (secondary lane membership, primary category Security / SOC agent)

    Security hyperautomation platform whose HyperSOC agents run triage, investigation and response inside governed workflows. Customers connect their own OpenAI, Azure OpenAI, Vertex AI, Anthropic or Bedrock subscription and pick the model per AI Agent and per AI Task, under roles, SSO, approval steps and audit logs that export to Amazon S3. It qualifies here on secondary lane membership rather than as a primary platform.

40. 40.Wonderful

    12.5 / 14 capabilities

    enterprise operations agents, agent builders (secondary lane membership, primary category Customer support agent)

    Nothing built is locked to one model: the platform routes each task to the model that fits, and new models drop in without rework. Admins use the AI Gateway to set which roles or teams can reach which models, cap budgets and limit premium model use. A model picker for each agent is not shown publicly.

The remaining 173 platforms that clear the bar

Every one of these documents model flexibility and routing in full. They score below 12.5 of 14 on total coverage, which says something about breadth across the whole taxonomy rather than about how open they are on models. Several of the strongest model positions in the index are narrow runtimes and gateways that score low precisely because they do one thing.

Common questions

Which AI agent platforms let you bring your own model or route between providers?

The Agentic Index rates 554 agentic AI platforms on model flexibility from each vendor's own public evidence. 213 of them document model flexibility and routing in full, meaning bring your own model, bring your own key, or routing tasks between models and providers. That is 38.4% of the platform field. The strongest documented positions by total capability coverage are Appian, FLOWX.AI, Gumloop, Mastra, Microsoft, ServiceNow and UiPath, followed by Pega and StackAI. Another 73 platforms document some model choice without enough evidence to clear, 260 document none, and 8 could not be verified either way. Graded from public evidence only.

Can you get model choice and enterprise governance in the same platform?

Yes, and the two go together more often than chance, but the combination is still uncommon. 82 of the 554 platforms in the pool document both model flexibility and the four capability enterprise governance bar of security and identity, deployment and residency, observability, and human approval before an agent acts. That is 14.8% of the field. The association is positive in both directions: platforms documenting model choice clear governance at 38.5% against a field base rate of 21.3%, and platforms clearing governance document model choice at 69.5% against a base rate of 38.4%. Each roughly doubles the odds of the other, and if they were independent the overlap would be about 45 platforms rather than 82. The scarcity is in the joint requirement, not in a conflict between the two.

What counts as bring your own model on this page?

Documented public evidence that model choice is a buyer facing surface: a provider list you can select from, bring your own model or bring your own key with your own provider account, or a routing policy that sends tasks between models. Partial evidence, meaning one alternative provider or a roadmap statement, does not clear. One honest limit worth knowing: the axis grades bring your own key and routing between providers as a single capability, so a Full grade does not by itself tell you which of the two a given platform documents. Read the entry and the linked vendor profile for that.

If a platform takes your own model, does it also run in your own environment?

More often than not. 171 of the 213 platforms here also document self hosted, on premises, air gapped or private VPC deployment in full, which is 80.3% against a pool base rate of 50.7%. The other 42 will take your model while the software keeps running in the vendor's cloud. That is a real answer for a team whose constraint is which model sees the data rather than where the software runs, and the wrong answer for a team whose constraint is the boundary.

Does a platform missing from this list mean it locks you to one model?

It means the vendor does not document model choice as a buyer facing option in its own public materials. Plenty of platforms swap models internally. The point is that a buyer could not verify it before signing, and could not choose after. Not documented is a statement about the evidence, not an accusation about the product.

Is this ranking paid or sponsored?

No. No vendor pays for placement, no vendor has reviewed this page, and every grade comes from the vendor's own public materials under the Agentic Index verification standard. 955 vendors are graded against the same 14 capabilities. Data last verified October 3, 2026.

Method: membership is the same 554 platform pool used by the best agentic AI platforms in 2026, drawn from 955 researched vendors. The editorial pages over this pool are one method with different bars, not several opinions. Every grade comes from the vendor's own public materials under the Agentic Index verification standard. No vendor pays for placement and no vendor has reviewed this page. Data last verified October 3, 2026. How this evidence is graded

Related: platforms you can self host or run in a private VPC, enterprise security and compliance platforms, platforms that support Model Context Protocol tools, how all 955 vendors score on model flexibility, with the rate in each of the 13 use case lanes, agent infrastructure platforms ranked on the production contract, compare platforms side by side.