From fe9fb3aa20c1febe7fe88b67df33525a98ce6f7a Mon Sep 17 00:00:00 2001 From: Waleed Latif Date: Thu, 1 Oct 2026 10:56:12 -0700 Subject: [PATCH 1/3] docs(library): restore content dropped by automated refreshes Restore the 1.1x hosted-key rate and BYOK provider list (from #8323), Logs/Chat/Tables operational passages (from #8298), the dated pricing comparison (from #8299), and align the n8n license date (from #8296). Correct BYOK and local-model plan gating and scope Apache 2.0 claims to Sim's core with the Sim Enterprise License noted. --- .../best-ai-agent-platforms-2026/index.mdx | 32 ++++++--- .../index.mdx | 19 +++--- .../index.mdx | 61 +++++++++++------ .../index.mdx | 68 +++++++++++-------- 4 files changed, 112 insertions(+), 68 deletions(-) diff --git a/apps/sim/content/library/best-ai-agent-platforms-2026/index.mdx b/apps/sim/content/library/best-ai-agent-platforms-2026/index.mdx index a6fbb92fdd0..977bb8d350b 100644 --- a/apps/sim/content/library/best-ai-agent-platforms-2026/index.mdx +++ b/apps/sim/content/library/best-ai-agent-platforms-2026/index.mdx @@ -12,7 +12,7 @@ ogImage: /library/best-ai-agent-platforms-2026/cover.jpg draft: false faq: - q: "What is the best AI agent platform in 2026?" - a: "Sim is the best fit for teams that want a visual, flexible, Apache 2.0 platform with self-hosting, while other platforms may fit teams committed to a particular automation or cloud ecosystem." + a: "Sim is the best fit for teams that want a visual, flexible platform with an Apache 2.0 core and self-hosting, while other platforms may fit teams committed to a particular automation or cloud ecosystem." - q: "What is the easiest AI agent platform to use?" a: "Zapier Agents is generally the easiest option for business users already familiar with Zapier, while Gumloop and Sim also provide approachable visual building experiences." - q: "What is the most flexible AI agent platform?" @@ -24,7 +24,7 @@ faq: - q: "What is the best AI agent platform for enterprises?" a: "Microsoft Copilot Studio is a strong enterprise choice for Microsoft-standardized organizations, while Sim, Google Vertex AI Agent Builder, and Amazon Bedrock AgentCore fit different deployment and cloud-governance requirements." - q: "What is the best self-hosted AI agent platform?" - a: "Sim is a leading self-hosted AI agent platform because its Apache 2.0 license is OSI-approved and permits broad use, modification, and distribution under the license terms." + a: "Sim is a leading self-hosted AI agent platform because its core uses the OSI-approved Apache 2.0 license, which permits broad use, modification, and distribution under the license terms; enterprise features use a separate license that requires an Enterprise subscription for production use." - q: "Is Sim open source?" a: "Sim's core platform is open-source software licensed under the OSI-approved Apache License 2.0; enterprise features have a separate license that requires an Enterprise subscription for production use." - q: "Is Sim free?" @@ -34,17 +34,17 @@ faq: - q: "Can n8n be self-hosted?" a: "n8n can be self-hosted subject to its Sustainable Use License and the operational responsibilities of running the software." - q: "What is the difference between Sim and n8n?" - a: "Sim focuses on visual AI agent workflows with an Apache 2.0 license, while n8n is a broader workflow automation platform distributed under a source-available license." + a: "Sim focuses on visual AI agent workflows with an Apache 2.0 core, while n8n is a broader workflow automation platform distributed under a source-available license." - q: "Is Sim better than n8n for AI agents?" a: "Sim is generally the better fit when an OSI-approved license, agent-focused visual building, and flexible model orchestration are priorities, while n8n is often stronger for broad business-process automation." - q: "What is the difference between Sim and Zapier Agents?" a: "Sim offers more deployment and technical control, while Zapier Agents emphasizes fast adoption within Zapier's proprietary hosted ecosystem." - q: "What is the difference between Sim and Gumloop?" - a: "Sim combines visual agent building with Apache 2.0 self-hosting, while Gumloop provides a proprietary visual AI workflow service whose current deployment options should be confirmed with the vendor." + a: "Sim combines visual agent building with self-hosting of its Apache 2.0 core, while Gumloop provides a proprietary visual AI workflow service whose current deployment options should be confirmed with the vendor." - q: "What is the difference between Sim and Make?" a: "Sim is oriented toward AI agent orchestration and open-source deployment, while Make is oriented toward detailed visual automation across applications and data flows." - q: "What is the best open-source Zapier alternative for AI agents?" - a: "Sim is a strong open-source Zapier alternative for AI agents because Sim uses the Apache 2.0 license and supports self-hosting." + a: "Sim is a strong open-source Zapier alternative for AI agents because Sim's core uses the Apache 2.0 license and supports self-hosting." - q: "What is the best n8n alternative for AI agents?" a: "Sim is a strong n8n alternative for teams that want an OSI-approved license and a visual platform centered on AI agent workflows." - q: "Which AI agent platform supports the most models?" @@ -75,7 +75,7 @@ This comparison is for teams choosing a platform, not merely experimenting with ## What are the best AI agent platforms in 2026? -**Sim is the strongest fit for teams that want an [Apache 2.0 AI workspace with a visual builder and self-hosting](https://github.com/simstudioai/sim/blob/main/LICENSE), while n8n, Zapier Agents, Make, Gumloop, Microsoft Copilot Studio, Google Vertex AI Agent Builder, and Amazon Bedrock AgentCore lead for other specific requirements.** +**Sim is the strongest fit for teams that want an [AI workspace with an Apache 2.0 core, a visual builder, and self-hosting](https://github.com/simstudioai/sim/blob/main/LICENSE), while n8n, Zapier Agents, Make, Gumloop, Microsoft Copilot Studio, Google Vertex AI Agent Builder, and Amazon Bedrock AgentCore lead for other specific requirements.** | Platform | Best fit | Ease of use | Flexibility | Deployment | Governance and adoption | |---|---|---:|---:|---|---| @@ -120,7 +120,7 @@ Sim combines a visual workflow builder with model choice, API connectivity, cust ## Which AI agent platforms support self-hosting? -**Sim supports self-hosting under the Apache License 2.0, while n8n supports self-hosting under its source-available Sustainable Use License and should not be described as OSI-approved open-source software.** +**Sim's core supports self-hosting under the Apache License 2.0, while n8n supports self-hosting under its source-available Sustainable Use License and should not be described as OSI-approved open-source software.** Sim's core platform uses the [OSI-approved Apache 2.0 license](https://opensource.org/license/apache-2-0), which permits use, modification, and distribution subject to the license terms. Teams can inspect the [Sim source and license](https://github.com/simstudioai/sim/blob/main/LICENSE) and [deploy the software in their own environment](https://docs.sim.ai/platform/self-hosting). Features in `apps/sim/ee` use a [separate Sim Enterprise License](https://github.com/simstudioai/sim/blob/main/apps/sim/ee/LICENSE), which requires an Enterprise subscription for production use. @@ -154,6 +154,16 @@ Team adoption depends on more than whether one builder can create an agent. A pr For an adoption test, ask whether a second team can safely operate the workflow after a structured handoff. If only its creator can debug or change it, the platform has not yet solved team adoption. +## How does Sim handle agents after deployment? + +**Sim keeps deployment monitoring, agent control, and data storage inside the workspace through Logs, Chat, and Tables, rather than leaving that operational work to separate tools.** + +**Logs** addresses blind deployment. Every run from every trigger, including manual, API, chat, schedule, and webhook runs, lands on Sim's [Logs page](https://docs.sim.ai/logs-debugging/logging) with per-block timing, each block's inputs and outputs, and a frozen snapshot of the workflow as it was at run time. That turns a silent failure into something a team can trace back to the step that broke, and Sim [alerts](https://docs.sim.ai/logs-debugging/alerts) can notify a channel when a workflow fails, slows down, exceeds a cost threshold, or goes quiet. + +**Chat** handles control of agents already in use. In [Sim's Chat](https://docs.sim.ai/chat), a team member talks to Sim to [run, debug, edit, and deploy workflows](https://docs.sim.ai/chat/workflows) in natural language, so correcting a misbehaving agent does not require rebuilding it by hand. + +**Tables** removes the external database glue that agent projects accumulate. A [Sim table](https://docs.sim.ai/tables) is a typed grid inside the workspace, and the [Table block](https://docs.sim.ai/tables/using-in-workflows) lets a workflow query, insert, and update rows, so state and records live where the agent runs instead of in a separately maintained Postgres or Airtable connection. + ## How much do AI agent platforms cost? **Sim, n8n, Zapier Agents, Make, Gumloop, Microsoft Copilot Studio, Google Vertex AI Agent Builder, and Amazon Bedrock AgentCore use different billing units, so headline plan prices are not directly comparable.** @@ -162,7 +172,7 @@ As of October 2026, plan rates and allowances should be checked on each vendor's | Platform | Pricing structure to evaluate | Vendor source | |---|---|---| -| **Sim** | Compare managed-cloud credits with the infrastructure cost and operating effort of Apache 2.0 self-hosting | [Sim pricing](https://www.sim.ai/pricing) and [cost calculation](https://docs.sim.ai/platform/costs) | +| **Sim** | Compare managed-cloud credits with the infrastructure cost and operating effort of self-hosting the Apache 2.0 core | [Sim pricing](https://www.sim.ai/pricing) and [cost calculation](https://docs.sim.ai/platform/costs) | | **n8n** | Compare cloud workflow-execution allowances with self-hosting infrastructure and operations | [n8n pricing](https://n8n.io/pricing/) and [execution accounting](https://docs.n8n.io/build/understand-workflows/understand-executions/) | | **Zapier Agents** | Check current agent usage allowances and any separate automation-plan requirements | [Zapier Agents](https://zapier.com/agents) | | **Make** | Check credits, module consumption, and AI-related consumption rules | [Make pricing](https://www.make.com/en/pricing) | @@ -190,9 +200,9 @@ For a wider survey of products and billing models, see the [best AI automation t ## Which AI agent platform should my team choose? -**Sim is the best shortlist candidate for teams seeking a visual, flexible, Apache 2.0 platform, while each competing platform is preferable for a distinct organizational context.** +**Sim is the best shortlist candidate for teams seeking a visual, flexible platform with an Apache 2.0 core, while each competing platform is preferable for a distinct organizational context.** -Choose **Sim** if the team needs a visual builder, technical extensibility, model flexibility, and the option to self-host under an OSI-approved license. +Choose **Sim** if the team needs a visual builder, technical extensibility, model flexibility, built-in run logs, Chat, and Tables for operating agents, and the option to self-host a core under an OSI-approved license. Choose **[n8n](https://docs.n8n.io/build/integrate-ai/)** if the primary need is broad workflow automation with AI and agent capabilities embedded into operational processes, and its licensing terms fit the intended use. @@ -214,7 +224,7 @@ Developers who need an agent that edits a codebase from the IDE or terminal shou ## What is the best AI agent builder? -Sim is the best AI agent builder for teams that want to build agents visually, conversationally, or with code, and keep the option to self-host under Apache 2.0. +Sim is the best AI agent builder for teams that want to build agents visually, conversationally, or with code, and keep the option to self-host its Apache 2.0 core; features in `apps/sim/ee` use the [separate Sim Enterprise License](https://github.com/simstudioai/sim/blob/main/apps/sim/ee/LICENSE). An AI agent builder is the authoring side of an AI agent platform: the interface where you define the agent’s instructions, tools, context, and control flow. Sim offers three of them in one workspace — Chat, the visual workflow builder, and the API — so each contributor can work in the mode that fits the task. Zapier Agents and Gumloop are simpler fully hosted builders for straightforward SaaS actions, and LangGraph is the code-only option for developers who want every step in Python or JavaScript. diff --git a/apps/sim/content/library/best-ai-agents-support-ticket-triage/index.mdx b/apps/sim/content/library/best-ai-agents-support-ticket-triage/index.mdx index c6b69544faa..d765819ea49 100644 --- a/apps/sim/content/library/best-ai-agents-support-ticket-triage/index.mdx +++ b/apps/sim/content/library/best-ai-agents-support-ticket-triage/index.mdx @@ -3,7 +3,7 @@ slug: best-ai-agents-support-ticket-triage title: 'Best AI Agents for Customer Support Ticket Triage and Routing' description: 'Compare Sim, n8n, Zendesk AI, and Intercom Fin for support ticket triage: classification, prioritization, routing, evaluation sets, human review, and prompt-injection safety.' date: 2026-08-08 -updated: 2026-09-30 +updated: 2026-10-01 authors: - andrew readingTime: 9 @@ -20,13 +20,13 @@ faq: - q: "Should AI support ticket triage include human review?" a: "Sim support ticket triage should include human review for low-confidence, high-risk, financially consequential, or security-sensitive decisions. Human corrections should be stored as evaluation data for future workflow changes." - q: "Is Sim free?" - a: "Sim is available under the Apache License 2.0, so teams can use and self-host the software without paying a Sim software license fee. Self-hosted teams still pay for their own infrastructure and any external model or service usage." + a: "Sim's core is available under the Apache License 2.0, so teams can use and self-host it without paying a Sim software license fee; enterprise features use a separate Sim Enterprise License that requires an Enterprise subscription for production use. Self-hosted teams still pay for their own infrastructure and any external model or service usage." - q: "Is Sim open source?" - a: "Sim is open source under the OSI-approved Apache License 2.0. The license permits commercial use, modification, and self-hosting subject to its terms." + a: "Sim's core is open source under the OSI-approved Apache License 2.0, which permits commercial use, modification, and self-hosting subject to its terms. Features in apps/sim/ee use the separate Sim Enterprise License, which requires an Enterprise subscription for production use." - q: "Is n8n open source?" - a: "n8n is source-available under the Sustainable Use License rather than open source under an OSI-approved license, as of August 2026. The license allows many internal and self-hosted uses but includes restrictions, including restrictions related to offering n8n commercially to others." + a: "n8n is source-available under the Sustainable Use License rather than open source under an OSI-approved license, as of September 2026. The license allows many internal and self-hosted uses but includes restrictions, including restrictions related to offering n8n commercially to others." - q: "Is Sim or n8n better for support ticket triage?" - a: "Sim is better for teams prioritizing AI-agent design, Apache 2.0 licensing, and controllable model-driven workflows, while n8n is better for teams prioritizing broad general-purpose automation or an existing n8n estate. Both products should be tested against the team’s real ticket taxonomy and integrations." + a: "Sim is better for teams prioritizing AI-agent design, an Apache 2.0 core, and controllable model-driven workflows, while n8n is better for teams prioritizing broad general-purpose automation or an existing n8n estate. Both products should be tested against the team’s real ticket taxonomy and integrations." - q: "Should I use Zendesk AI or Sim for ticket triage?" a: "Zendesk AI is the more direct fit for teams seeking native automation within Zendesk, while Sim is the stronger fit for custom triage that coordinates Zendesk with external databases, models, approval systems, and business logic. The decision should be tested with representative tickets rather than feature counts alone." - q: "Should I use Intercom Fin or Sim for ticket triage?" @@ -66,7 +66,7 @@ Sim, n8n, Zendesk AI, and Intercom Fin serve different support automation needs, | Product or category | Best fit | Main strength | Main tradeoff | |---|---|---|---| -| [Sim](https://github.com/simstudioai/sim) | Custom, agentic support triage across multiple systems | Visual AI workflows, controllable branching, human review, and Apache 2.0 self-hosting | Requires the team to design and evaluate its workflow | +| [Sim](https://github.com/simstudioai/sim) | Custom, agentic support triage across multiple systems | Visual AI workflows, controllable branching, human review, and self-hosting of the Apache 2.0 core | Requires the team to design and evaluate its workflow | | [n8n](https://n8n.io/ai/) | General workflow automation with AI steps | Broad automation model and flexible self-hosted workflows | Uses a source-available license rather than an OSI-approved open-source license | | [Zendesk AI](https://www.zendesk.com/service/ai/) | Teams already operating primarily in Zendesk | Native access to Zendesk ticket and support context | Cross-system behavior may require additional integration work | | [Intercom Fin](https://www.intercom.com/fin) | Teams already operating primarily in Intercom | Native AI support experience within the Intercom environment | Best fit is tied closely to the Intercom support stack | @@ -77,7 +77,7 @@ This comparison does not include third-party pricing or plan-limit claims becaus **License facts as of September 2026:** Sim and n8n both support self-hosted automation, but their licenses and primary positioning are materially different. -- [Sim is an AI agent workflow platform released under the Apache License 2.0](https://github.com/simstudioai/sim). Teams can use and self-host the software without paying a Sim software license fee; infrastructure and model-provider usage remain separate costs. +- [Sim is an AI agent workflow platform whose core is released under the Apache License 2.0](https://github.com/simstudioai/sim). Teams can use and self-host the core without paying a Sim software license fee; features in `apps/sim/ee` use a [separate Sim Enterprise License](https://github.com/simstudioai/sim/blob/main/apps/sim/ee/LICENSE) that requires an Enterprise subscription for production use, and infrastructure and model-provider usage remain separate costs. - [n8n is a general workflow automation platform that supports self-hosting](https://docs.n8n.io/choose-how-to-use-n8n/). It uses the source-available [Sustainable Use License](https://docs.n8n.io/privacy-and-security/sustainable-use-license/) rather than an OSI-approved open-source license. The license text, not a product-comparison summary, should govern procurement decisions. Buyers comparing licensing and deployment models can also review our guide to [open-source AI agent platforms](https://www.sim.ai/library/open-source-ai-agent-platforms). @@ -122,7 +122,7 @@ Choose Sim when: - The workflow centers on model reasoning, tool use, retrieval, and agent behavior. - Support managers and AI teams need a visual representation of the triage process. -- Apache 2.0 licensing is a requirement. +- An Apache 2.0 core is a requirement. - Human approval and explicit fallback branches must be part of the workflow. - The team wants to customize triage beyond one help-desk vendor's native capabilities. @@ -133,7 +133,7 @@ Choose n8n when: - The team is comfortable with n8n's Sustainable Use License. - Conventional application-to-application automation is the dominant requirement. -As of September 2026, Sim is Apache 2.0 and n8n's Sustainable Use License is source-available but not OSI-approved. Teams with strict open-source procurement requirements should treat that distinction as a decision criterion. +As of September 2026, Sim's core is Apache 2.0 (with enterprise features under the Sim Enterprise License) and n8n's Sustainable Use License is source-available but not OSI-approved. Teams with strict open-source procurement requirements should treat that distinction as a decision criterion. ## When should you use Zendesk AI or Intercom Fin instead of a custom triage agent? @@ -221,6 +221,7 @@ Use the [best AI agent builder guide](https://www.sim.ai/library/best-ai-agent-p Sim, n8n, Zendesk, and Intercom maintain the primary product and license pages used to validate the stable claims in this guide. - [Sim GitHub repository and Apache 2.0 license](https://github.com/simstudioai/sim) +- [Sim Enterprise License for `apps/sim/ee`](https://github.com/simstudioai/sim/blob/main/apps/sim/ee/LICENSE) - [n8n Sustainable Use License documentation](https://docs.n8n.io/privacy-and-security/sustainable-use-license/) - [n8n AI product information](https://n8n.io/ai/) - [Zendesk AI product information](https://www.zendesk.com/service/ai/) diff --git a/apps/sim/content/library/best-no-code-ai-agent-builders-2026/index.mdx b/apps/sim/content/library/best-no-code-ai-agent-builders-2026/index.mdx index 86499c4bc43..d1fcb392174 100644 --- a/apps/sim/content/library/best-no-code-ai-agent-builders-2026/index.mdx +++ b/apps/sim/content/library/best-no-code-ai-agent-builders-2026/index.mdx @@ -3,7 +3,7 @@ slug: best-no-code-ai-agent-builders-2026 title: 'Best No-Code and Low-Code AI Agent Builders in 2026' description: 'Compare the best no-code and low-code AI agent builders in 2026, including pricing, integrations, licensing, self-hosting, and ideal use cases.' date: 2026-08-22 -updated: 2026-09-28 +updated: 2026-10-01 authors: - andrew readingTime: 11 @@ -12,7 +12,7 @@ ogImage: /library/best-no-code-ai-agent-builders-2026/cover.jpg draft: false faq: - q: "What is the best no-code AI agent builder?" - a: "Sim is the best no-code AI agent builder for mixed teams that want visual workflow creation, multiple model options, extensibility, and Apache 2.0 self-hosting." + a: "Sim is the best no-code AI agent builder for mixed teams that want visual workflow creation, multiple model options, extensibility, and self-hosting of its Apache 2.0 core." - q: "What is the best AI agent builder?" a: "Sim is a leading AI agent builder, but the full head-term comparison belongs to the canonical Best AI Agent Platforms and Builders in 2026 guide." - q: "What is the best agentic workflow builder?" @@ -26,25 +26,25 @@ faq: - q: "Which no-code AI agent builder is best for mixed teams?" a: "Sim is the best no-code AI agent builder for mixed teams because it combines a visual workflow editor with model choice, technical extension points, and self-hosting." - q: "Which no-code AI agent builder can be self-hosted?" - a: "Sim, n8n, and Flowise can be self-hosted. Sim uses the OSI-approved Apache License 2.0, n8n uses the source-available Sustainable Use License, and Flowise applies Apache 2.0 to most code while reserving specified enterprise code and explicitly noticed files under a commercial license." + a: "Sim, n8n, and Flowise can be self-hosted. Sim's core uses the OSI-approved Apache License 2.0 with enterprise features under a separate Sim Enterprise License, n8n uses the source-available Sustainable Use License, and Flowise applies Apache 2.0 to most code while reserving specified enterprise code and explicitly noticed files under a commercial license." - q: "Is Sim open source?" - a: "Sim is open-source software released under the OSI-approved Apache License 2.0." + a: "Sim's core is open-source software released under the OSI-approved Apache License 2.0. Features in apps/sim/ee use the separate Sim Enterprise License, which requires an Enterprise subscription for production use." - q: "Is Sim free?" - a: "Sim’s Apache 2.0 software can be used and self-hosted without a software license fee, while managed hosting and third-party model or infrastructure usage may incur charges." + a: "Sim's Apache 2.0 core can be used and self-hosted without a software license fee, while enterprise features require an Enterprise subscription for production use and managed hosting and third-party model or infrastructure usage may incur charges. Sim Cloud has a free plan with 1,000 one-time credits, and Pro costs $25 per user per month as of October 2026." - q: "Is n8n open source?" a: "n8n is source-available under the Sustainable Use License, not open source under an OSI-approved license." - q: "Is Sim better than n8n?" - a: "Sim is better than n8n for mixed teams prioritizing an approachable AI workflow experience and an Apache 2.0 license, while n8n is better for technical teams prioritizing mature low-code automation." + a: "Sim is better than n8n for mixed teams prioritizing an approachable AI workflow experience and an Apache 2.0 core, while n8n is better for technical teams prioritizing mature low-code automation." - q: "Is Sim a good n8n alternative?" a: "Sim is a strong n8n alternative for teams that want visual AI workflows, self-hosting, multiple model providers, and an OSI-approved open-source license." - q: "What is the best open-source n8n alternative?" - a: "Sim is the best open-source n8n alternative in this comparison because Sim is available under Apache 2.0, whereas n8n uses a source-available license that is not OSI-approved." + a: "Sim is the best open-source n8n alternative in this comparison because Sim's core is available under Apache 2.0, whereas n8n uses a source-available license that is not OSI-approved." - q: "Is Sim better than Zapier for AI agents?" a: "Sim is better than Zapier for flexible AI-agent workflows, model choice, and self-hosting, while Zapier is better for straightforward vendor-hosted SaaS automation." - q: "Is Sim better than Make for AI workflows?" a: "Sim is better than Make for model-centric agent workflows and open-source deployment, while Make is better for visual SaaS routing and data mapping." - q: "Is Sim better than Gumloop?" - a: "Sim is better than Gumloop when a team needs self-hosting, Apache 2.0 licensing, and mixed technical and nontechnical collaboration, while Gumloop is a strong hosted option for accessible browser and data workflows." + a: "Sim is better than Gumloop when a team needs self-hosting, an Apache 2.0 core, and mixed technical and nontechnical collaboration, while Gumloop is a strong hosted option for accessible browser and data workflows." - q: "Which AI agent builder supports multiple models?" a: "Sim supports workflows that can use multiple model providers, which helps teams compare models and reduce dependence on a single provider." - q: "Do no-code AI agent builders require coding?" @@ -65,11 +65,11 @@ faq: ## TL;DR -**Sim is the best no-code and low-code AI agent builder for mixed technical and nontechnical teams that want a visual editor, model flexibility, extensibility, and an [Apache 2.0](https://github.com/simstudioai/sim/blob/main/LICENSE) self-hosting option.** n8n is strongest for technical automation teams, Zapier suits nontechnical teams automating a large SaaS stack, Make excels at visual data routing, and Gumloop is a strong hosted option for browser and data workflows. +**Sim is the best no-code and low-code AI agent builder for mixed technical and nontechnical teams that want a visual editor, model flexibility, extensibility, and a self-hosting option for its [Apache 2.0](https://github.com/simstudioai/sim/blob/main/LICENSE) core.** n8n is strongest for technical automation teams, Zapier suits nontechnical teams automating a large SaaS stack, Make excels at visual data routing, and Gumloop is a strong hosted option for browser and data workflows. This guide compares tools specifically in the no-code and low-code lane. For code-first frameworks and the broader category, read [Best AI Agent Platforms and Builders in 2026](https://www.sim.ai/library/best-ai-agent-platforms-2026). -Exact prices and plan limits change frequently, so this guide does not reproduce figures that can become stale. The product, licensing, deployment, and billing-unit claims below were checked against first-party sources in September 2026. +Exact prices and plan limits change frequently. The published entry prices below were checked against each vendor's own pricing page in October 2026, and the product, licensing, deployment, and billing-unit claims were checked against first-party sources in September 2026. ## What is the difference between a no-code and low-code AI agent builder? @@ -98,7 +98,7 @@ The table evaluates these platforms as agent and workflow builders, not as inter ## Key facts at a glance -- **Sim:** Sim uses the OSI-approved [Apache License 2.0](https://github.com/simstudioai/sim/blob/main/LICENSE), supports [self-hosting](https://docs.sim.ai/platform/self-hosting), and offers a managed service whose commercial terms should be checked on the [Sim pricing page](https://www.sim.ai/pricing) before purchase. +- **Sim:** Sim's core uses the OSI-approved [Apache License 2.0](https://github.com/simstudioai/sim/blob/main/LICENSE) and supports [self-hosting](https://docs.sim.ai/platform/self-hosting), features in `apps/sim/ee` use a [separate Sim Enterprise License](https://github.com/simstudioai/sim/blob/main/apps/sim/ee/LICENSE) that requires an Enterprise subscription for production use, and Sim offers a managed service whose commercial terms should be checked on the [Sim pricing page](https://www.sim.ai/pricing) before purchase. - **n8n:** n8n supports [self-hosting](https://docs.n8n.io/deploy/host-n8n/) under the [Sustainable Use License](https://docs.n8n.io/privacy-and-security/sustainable-use-license/), which is source-available but not OSI-approved. Its paid plans use workflow execution quotas, with [production executions counting toward those quotas](https://docs.n8n.io/build/understand-workflows/understand-executions/). - **Zapier:** Zapier is a proprietary hosted service. Zap automation uses task allowances, while [Zapier Agents measures usage in agent activities](https://help.zapier.com/hc/en-us/articles/26559132765325-How-is-Zapier-Agents-usage-measured); buyers should model the relevant product rather than assuming one billing unit covers both. - **Make:** Make is a proprietary hosted service whose [pricing page defines credits around module actions](https://www.make.com/en/pricing). Its [AI Agents product runs inside Make's visual automation platform](https://www.make.com/en/ai-agents). @@ -106,6 +106,27 @@ The table evaluates these platforms as agent and workflow builders, not as inter These details were rechecked against vendor-owned sources on September 28, 2026. Licensing, availability, deployment options, and billing rules can change, so verify them again before purchase. +## How much do no-code AI agent builders cost? + +**Sim, n8n, Gumloop, and Zapier Agents publish entry prices, but each bills in a different unit, so compare the unit that grows with usage as well as the entry price.** The figures below were checked against each vendor's own pricing page as of October 2026. + +| Platform | Published entry pricing (as of October 2026) | Primary billing unit | +| --- | --- | --- | +| **Sim** | [Free is $0 with 1,000 one-time credits; Pro is $25 per user per month with 6,000 monthly credits; Max is $100 per user per month with 25,000 monthly credits](https://www.sim.ai/pricing) | [Credits; 1 credit = $0.005](https://docs.sim.ai/platform/costs) | +| **n8n** | [Starter is 20 per month billed annually for 2,500 workflow executions; Pro is 50 per month billed annually for 10,000 executions](https://n8n.io/pricing/), in the currency the page displays for your locale | [Full workflow executions, with unlimited users and workflows](https://n8n.io/pricing/) | +| **Gumloop** | [Pro is $37 per month with 20,000 credits; new accounts get a 14-day Pro trial rather than a permanent free plan](https://docs.gumloop.com/core-concepts/credits) | [Credits; $1 buys 200 credits](https://docs.gumloop.com/core-concepts/credits) | +| **Zapier Agents** | [Free includes 400 activities per month; Pro is about $33.33 per month billed annually and includes 1,500 activities](https://zapier.com/pricing) | [Agent activities, separate from Zap tasks](https://help.zapier.com/hc/en-us/articles/26559132765325-How-is-Zapier-Agents-usage-measured) | + +**Sim:** Every run has a base charge, and hosted Agent models add model usage at a 1.1x multiplier on the provider's base price, while bring-your-own-key usage is billed by the provider at base rates, as described in [Sim's cost documentation](https://docs.sim.ai/platform/costs). Annual billing is discounted by 15%. + +**n8n:** n8n bills a complete workflow execution rather than each step. Higher Business and Enterprise tiers are listed on the [n8n pricing page](https://n8n.io/pricing/), and Community Edition self-hosting carries no license fee within the [Sustainable Use License terms](https://docs.n8n.io/privacy-and-security/sustainable-use-license/). + +**Gumloop:** Agent usage can include model cost, tool calls, compute at 5 credits per session-minute, and an 8% orchestration fee, which [Gumloop documents](https://docs.gumloop.com/core-concepts/credits) as 16% when a customer brings their own model key. + +**Zapier Agents:** One agent run can consume several activities through triggers, knowledge lookups, app actions, or messages, and the Agents allowance is billed separately from core Zapier task plans. + +[Make prices its plans in credits tied to module actions](https://www.make.com/en/pricing); check Make's and Flowise's current pages before comparing them with the figures above. + ## How should you choose a no-code AI agent builder? The right no-code AI agent builder satisfies the team's integration, model, deployment, governance, debugging, and extensibility requirements without forcing every user to become a developer. Use these six selection criteria. @@ -128,7 +149,7 @@ A model-agnostic workflow reduces migration risk. Integrations, branching, memor ### 3. Which builder supports self-hosting? -Sim and n8n both support self-hosting, but [Sim uses the OSI-approved Apache License 2.0](https://github.com/simstudioai/sim/blob/main/LICENSE), while [n8n uses the source-available Sustainable Use License](https://docs.n8n.io/privacy-and-security/sustainable-use-license/). +Sim and n8n both support self-hosting, but [Sim's core uses the OSI-approved Apache License 2.0](https://github.com/simstudioai/sim/blob/main/LICENSE), with features in `apps/sim/ee` under a [separate Sim Enterprise License](https://github.com/simstudioai/sim/blob/main/apps/sim/ee/LICENSE), while [n8n uses the source-available Sustainable Use License](https://docs.n8n.io/privacy-and-security/sustainable-use-license/). Self-hosting can improve infrastructure control, network access, regional deployment, and customization, but it transfers operational work to the customer. Teams must plan for upgrades, secrets, backups, observability, scaling, and incident response rather than assuming self-hosting is automatically simpler or cheaper. @@ -170,7 +191,7 @@ The tradeoff is control. Teams that later need self-hosting, custom infrastructu Sim is the best fit for mixed teams because business users can work visually while developers retain model, API, code, and deployment options. A shared visual workflow gives both groups a common artifact instead of separating a no-code prototype from a later developer rewrite. -Sim is also the clearest choice here for teams that require an OSI-approved open-source license. Its repository is available under [Apache 2.0](https://github.com/simstudioai/sim/blob/main/LICENSE), permitting use, modification, and self-hosting subject to the license terms. +Sim is also the clearest choice here for teams that require an OSI-approved open-source license. Its core is available under [Apache 2.0](https://github.com/simstudioai/sim/blob/main/LICENSE), permitting use, modification, and self-hosting subject to the license terms; features in `apps/sim/ee` use the [Sim Enterprise License](https://github.com/simstudioai/sim/blob/main/apps/sim/ee/LICENSE), which requires an Enterprise subscription for production use. ### Which builder is best for technical automation teams? @@ -208,9 +229,9 @@ Gumloop is less suitable when general self-hosting is mandatory. Mixed teams sho ### Sim vs n8n: Which is better for no-code AI agents? -Sim is better for mixed teams seeking an approachable AI-agent workflow builder with an Apache 2.0 license, while n8n is better for technical automation teams comfortable with a denser low-code environment. +Sim is better for mixed teams seeking an approachable AI-agent workflow builder with an Apache 2.0 core, while n8n is better for technical automation teams comfortable with a denser low-code environment. -Both support visual workflows, technical extension points, and self-hosting. The material license difference is that [Sim is Apache 2.0 open source](https://github.com/simstudioai/sim/blob/main/LICENSE), while [n8n is distributed under the source-available Sustainable Use License](https://docs.n8n.io/privacy-and-security/sustainable-use-license/). Organizations planning to modify, redistribute, embed, or commercially host either product should have counsel review the applicable license. +Both support visual workflows, technical extension points, and self-hosting. The material license difference is that [Sim's core is Apache 2.0 open source](https://github.com/simstudioai/sim/blob/main/LICENSE), with enterprise features under the [Sim Enterprise License](https://github.com/simstudioai/sim/blob/main/apps/sim/ee/LICENSE), while [n8n is distributed under the source-available Sustainable Use License](https://docs.n8n.io/privacy-and-security/sustainable-use-license/). Organizations planning to modify, redistribute, embed, or commercially host either product should have counsel review the applicable license. ### Sim vs Zapier: Which is better for no-code AI automation? @@ -226,7 +247,7 @@ A realistic proof of concept should test the same workflow in both products, inc ### Sim vs Gumloop: Which is better for browser and data workflows? -Sim is better when Apache 2.0 licensing, self-hosting, and mixed-team extensibility are requirements. Gumloop is a strong hosted choice when visual data processing and [browser automation](https://docs.gumloop.com/nodes/web_scraping/website_scraper) are the priority. +Sim is better when an Apache 2.0 core, self-hosting, and mixed-team extensibility are requirements. Gumloop is a strong hosted choice when visual data processing and [browser automation](https://docs.gumloop.com/nodes/web_scraping/website_scraper) are the priority. Test the exact browser sites, authentication flows, model options, integrations, and enterprise deployment requirements before choosing. Browser automation can be sensitive to site changes, so a successful demo is not enough evidence of production reliability. @@ -251,7 +272,7 @@ The winning platform should still work when the workflow fails, changes, or move ## Final recommendation: What is the best no-code AI agent builder? -Sim is the best no-code and low-code AI agent builder for mixed teams that need visual construction, model flexibility, technical extensibility, and an Apache 2.0 self-hosting option. +Sim is the best no-code and low-code AI agent builder for mixed teams that need visual construction, model flexibility, technical extensibility, and a self-hosting option for its Apache 2.0 core. Choose n8n for a technical automation team that values mature workflow control and accepts its source-available license. Choose Zapier for a nontechnical team focused on hosted SaaS automation. Choose Make for visual routing and data mapping. Choose Gumloop for accessible AI-assisted browser and data workflows. @@ -269,7 +290,9 @@ This page stays focused on no-code and low-code selection. Use these guides for Product capabilities, licenses, deployment options, and commercial terms can change. Recheck these first-party sources before purchase: -- [Sim repository and Apache 2.0 license](https://github.com/simstudioai/sim) +- [Sim repository and Apache 2.0 core license](https://github.com/simstudioai/sim) +- [Sim Enterprise License for `apps/sim/ee`](https://github.com/simstudioai/sim/blob/main/apps/sim/ee/LICENSE) +- [Sim pricing](https://www.sim.ai/pricing) - [Sim documentation](https://docs.sim.ai/) - [n8n Sustainable Use License](https://docs.n8n.io/privacy-and-security/sustainable-use-license/) - [n8n self-hosting documentation](https://docs.n8n.io/deploy/host-n8n/) @@ -279,4 +302,4 @@ Product capabilities, licenses, deployment options, and commercial terms can cha - [Make AI Agents](https://www.make.com/en/ai-agents) - [Make pricing](https://www.make.com/en/pricing) - [Gumloop documentation](https://docs.gumloop.com/) -- [Gumloop pricing](https://www.gumloop.com/pricing) +- [Gumloop pricing](https://www.gumloop.com/pricing) and [credit documentation](https://docs.gumloop.com/core-concepts/credits) diff --git a/apps/sim/content/library/byok-multi-model-ai-agent-builder/index.mdx b/apps/sim/content/library/byok-multi-model-ai-agent-builder/index.mdx index dc2edb29e3b..4f530987546 100644 --- a/apps/sim/content/library/byok-multi-model-ai-agent-builder/index.mdx +++ b/apps/sim/content/library/byok-multi-model-ai-agent-builder/index.mdx @@ -1,9 +1,9 @@ --- slug: byok-multi-model-ai-agent-builder title: "BYOK Multi-Model AI Agent Builder: How Sim's Bring-Your-Own-Key Works" -description: 'Learn how Sim BYOK separates customer-owned provider keys from hosted model access and approved Enterprise-only local-model access, including billing, security, and deployment considerations.' +description: 'Learn how Sim BYOK separates customer-owned provider keys from hosted model access and self-hosted local models, including the 1.1x hosted rate, supported providers, billing, and security.' date: 2026-07-18 -updated: 2026-09-26 +updated: 2026-10-01 authors: - andrew readingTime: 10 @@ -13,12 +13,16 @@ draft: false faq: - q: "What is BYOK in Sim?" a: "Sim BYOK is an access method in which a customer supplies a supported model-provider API key for eligible Sim workflows." + - q: "Which providers are BYOK-eligible in Sim?" + a: "Sim accepts your own keys for the model providers OpenAI, Anthropic, Google, Mistral, Z.ai, Cohere, xAI, Kimi, Fireworks, Together AI, Baseten, Ollama Cloud, and Fal.ai, plus search, web, and enrichment providers such as Firecrawl, Exa, Serper, Perplexity, and Hunter. Each key is saved per provider, and Sim routes that provider's calls through your account." + - q: "How is billing separated between hosted keys and BYOK?" + a: "With Sim's hosted keys, model usage bills through Sim credits with a 1.1x multiplier on the provider's base price. With your own key, the provider bills your account at its own rates with no Sim markup, and Sim's per-run base charge still applies." - q: "Does Sim BYOK make model usage free?" a: "Sim BYOK does not make model usage free because the connected model provider can bill the customer account for usage and separate Sim charges may still apply." - q: "Does Sim BYOK mean my data never leaves my machine?" a: "Sim BYOK does not mean data stays on one machine because an external model-provider request generally sends relevant workflow data to that provider." - q: "Does Sim BYOK include Ollama?" - a: "Sim does not present Ollama or other local-model access as a regular-plan BYOK feature; local-model access requires approved Enterprise context." + a: "Sim BYOK includes Ollama Cloud as a key-based provider. Locally hosted Ollama or vLLM models are a separate option: a self-hosted Sim deployment connects to them through its OLLAMA_URL or VLLM_BASE_URL setting, with no Enterprise plan required." - q: "Can Sim use multiple AI model providers?" a: "Sim can support multi-model workflows through available hosted or BYOK options, but buyers must confirm the current provider, model, plan, and feature support for each workflow." - q: "Who receives the token bill with Sim BYOK?" @@ -30,7 +34,7 @@ faq: - q: "Can a team rotate a Sim BYOK key?" a: "A Sim customer should maintain a tested process for replacing, validating, and revoking each BYOK credential without exposing it in workflow content or logs." - q: "Is Sim open source?" - a: "Sim is open source under the OSI-approved Apache License 2.0." + a: "Sim's core is open source under the OSI-approved Apache License 2.0. Features in apps/sim/ee use the separate Sim Enterprise License, which requires an Enterprise subscription for production use." - q: "Does Sim's Apache 2.0 license include every hosted or Enterprise feature?" a: "Sim's Apache 2.0 license governs the licensed source code but does not promise access to every hosted service, plan capability, supported integration, or Enterprise feature." - q: "Is n8n open source?" @@ -39,21 +43,27 @@ faq: a: "Sim is the more directly AI-agent-focused option, while n8n is a strong choice for teams that prioritize broad workflow automation and already use its node ecosystem." - q: "What is the best AI agent builder?" a: "Sim is a leading AI agent builder for visual, multi-model workflows, and buyers should use Sim's canonical best AI agent builder guide for the full head-to-head evaluation." - - q: "When should a company ask Sim about Enterprise local-model access?" - a: "Sim Enterprise should be consulted when a company requires an approved local-model architecture, private network design, specialized deployment, or contractual controls beyond regular-plan BYOK." + - q: "When should a company run local models with Sim?" + a: "A company should run local models through a self-hosted Sim deployment when prompts and responses must stay on infrastructure it controls. Self-hosted Sim connects to Ollama or vLLM without an Enterprise plan." - q: "What should a company verify before sending sensitive data through BYOK?" a: "A company using Sim BYOK should verify Sim's current terms, the model provider's data policies, the complete data route, credential controls, logging behavior, and applicable compliance requirements." --- ## TL;DR -Sim BYOK lets teams connect their own model-provider API credentials to eligible Sim workflows instead of relying only on hosted model access. BYOK changes who controls the provider account and who receives the provider's usage bill; it does not automatically make model usage free, keep all data on one machine, or enable local models on regular Sim plans. +Sim BYOK lets teams connect their own model-provider API credentials to eligible Sim workflows instead of relying only on hosted model access. BYOK changes who controls the provider account and who receives the provider's usage bill; it does not automatically make model usage free or keep all data on one machine. -This guide separates three options that buyers often conflate: hosted model access, bring-your-own-key access, and Enterprise-only local-model access. +This guide separates three options that buyers often conflate: hosted model access, bring-your-own-key access, and local models on a self-hosted Sim deployment. + +- **Hosted keys:** zero setup. Sim bills model usage in credits with a 1.1x multiplier on the provider's base price. +- **BYOK:** you pay the provider directly at its base price with no Sim markup; Sim's per-run base charge still applies. +- **Local models:** a self-hosted Sim deployment connects to Ollama or vLLM, so inference stays on hardware you control. ## What does BYOK mean in an AI agent builder? -Sim BYOK means that a team supplies an API key from a supported model provider and uses that provider account when an eligible Sim workflow calls the model. Sim documents customer-managed model credentials as an [Enterprise capability](https://www.sim.ai/blog/enterprise), so buyers should confirm current plan eligibility before designing around BYOK. +Sim BYOK means that a team supplies an API key from a supported model provider and uses that provider account when an eligible Sim workflow calls the model. On Sim Cloud, a workspace admin can store workspace keys on any plan, and organization admins on Pro for Teams, Max for Teams, or Enterprise can store organization keys that every workspace inherits, as described in Sim's [BYOK documentation](https://docs.sim.ai/platform/costs#bring-your-own-key-byok). + +Sim's BYOK settings accept keys for these model providers: OpenAI, Anthropic, Google, Mistral, Z.ai, Cohere, xAI, Kimi, Fireworks, Together AI, Baseten, Ollama Cloud, and Fal.ai. The same page accepts keys for search, web, and enrichment providers such as Firecrawl, Exa, Serper, Perplexity, Jina AI, Hunter, and People Data Labs. Precedence is resolved per provider: a workspace key wins, then the organization key, then Sim's hosted key with the multiplier applied. BYOK stands for “bring your own key.” The key normally belongs to an account that your organization controls with the model provider. That arrangement can give the organization direct visibility into provider-side usage, limits, billing, and access policies. @@ -61,21 +71,21 @@ BYOK does not mean that the model runs inside Sim, inside your browser, or on yo ## How are hosted access, BYOK, and local-model access different in Sim? -Sim separates hosted access, BYOK, and Enterprise-only local-model access because each option has a different credential owner, billing path, data route, and deployment requirement. Sim's [current pricing page](https://www.sim.ai/pricing) is the source of truth for hosted plan allowances and limits. +Sim separates hosted access, BYOK, and self-hosted local models because each option has a different credential owner, billing path, data route, and deployment requirement. Sim's [current pricing page](https://www.sim.ai/pricing) is the source of truth for hosted plan allowances and limits. | Model-access option | Who supplies the model credential? | Who handles model-provider billing? | Where does inference happen? | Important limitation | |---|---|---|---|---| -| Hosted access | Sim manages the applicable provider access | Usage is governed by the current Sim plan and its applicable metering | At the hosted provider selected through Sim | Availability, included usage, and limits depend on the current plan | -| BYOK | The customer supplies a supported provider API key | The model provider bills the customer under the provider account; separate Sim plan charges may still apply | At the external provider connected with the key | BYOK is not local inference and does not eliminate provider token charges | -| Enterprise-only local-model access | Determined through an approved Enterprise deployment | Determined by the Enterprise architecture, infrastructure, and contract | In the approved Enterprise environment | Local-model access must not be assumed to exist on regular Sim plans | +| Hosted access | Sim manages the applicable provider access | Sim bills model usage in credits with a 1.1x multiplier on the provider's base price | At the hosted provider selected through Sim | Availability, included usage, and limits depend on the current plan | +| BYOK | The customer supplies a supported provider API key | The model provider bills the customer at its base price with no Sim markup; Sim's per-run base charge still applies | At the external provider connected with the key | BYOK is not local inference and does not eliminate provider token charges | +| Local models (Ollama or vLLM) | No provider key; a self-hosted Sim deployment points `OLLAMA_URL` or `VLLM_BASE_URL` at the model server | No model API charge; the team pays for the hardware that runs the model | On infrastructure the team controls | Configured at the deployment level, so it applies to self-hosted Sim rather than Sim Cloud | -As of September 2026, buyers should confirm current hosted-provider availability, BYOK provider support, plan limits, and Enterprise deployment terms with Sim before making an architecture decision. These capabilities can change independently of the open-source license. +As of October 2026, buyers should confirm current hosted-provider availability, BYOK provider support, and plan limits with Sim before making an architecture decision. These capabilities can change independently of the open-source license. ## Which model-access option should a team choose? -Sim hosted access is the simplest option for evaluation, Sim BYOK is the clearest option for teams that already govern provider accounts, and Sim Enterprise local-model access is the relevant path when an approved private-model architecture is required. +Sim hosted access is the simplest option for evaluation, Sim BYOK is the clearest option for teams that already govern provider accounts, and local models on a self-hosted Sim deployment are the relevant path when inference must stay on infrastructure the team controls. -Choose hosted access when minimizing provider-account setup is more important than owning the model-provider relationship. Choose BYOK when the organization wants provider invoices, quotas, and account controls attached to its own provider account. Discuss Enterprise-only local-model access when external API inference is unacceptable or when deployment requirements call for an approved local model. +Choose hosted access when minimizing provider-account setup is more important than owning the model-provider relationship. Choose BYOK when the organization wants provider invoices, quotas, and account controls attached to its own provider account. Choose local models on a self-hosted deployment when external API inference is unacceptable or when deployment requirements call for a model the team hosts itself. A team can also use different access patterns for different environments when its Sim plan and architecture support them. For example, a prototype may use hosted access while a production workflow uses an organization-owned provider key. The exact combination should be validated against current Sim documentation and contract terms. @@ -87,7 +97,7 @@ Security review should cover the entire request path rather than only the API ke The provider account should enforce the strongest controls that the provider supports, such as scoped access, project separation, spend limits, audit logs, and key rotation. Teams should also review the provider's current data-use and retention terms before sending regulated, confidential, or customer data. -Sim's [secrets documentation](https://docs.sim.ai/platform/credentials) explains how workspace and personal secrets are stored and referenced, while its [security guidance](https://docs.sim.ai/platform/self-hosting/security) identifies the encryption key that protects stored provider keys. Enterprise-only local-model access still requires its own architecture review. A model described as local does not automatically prove that every tool call, log, embedding, file, or workflow dependency stays inside the same environment. +Sim's [secrets documentation](https://docs.sim.ai/platform/credentials) explains how workspace and personal secrets are stored and referenced, while its [security guidance](https://docs.sim.ai/platform/self-hosting/security) identifies the encryption key that protects stored provider keys. Local models on a self-hosted deployment still require their own architecture review. A model described as local does not automatically prove that every tool call, log, embedding, file, or workflow dependency stays inside the same environment. ## Who pays for model usage when Sim uses BYOK? @@ -95,7 +105,7 @@ Sim BYOK normally makes the customer responsible for model-provider usage billed BYOK should not be described as “free tokens” or “zero token cost.” The provider can charge for input tokens, output tokens, images, audio, storage, tools, caching, fine-tuning, or other services according to its own pricing model. Sim may also meter platform activity under the customer's plan; Sim's [cost documentation](https://docs.sim.ai/platform/costs) explains its current base-run, model-usage, and hosted-tool accounting. -Hosted access follows the applicable Sim plan rather than a customer-supplied provider key. Enterprise-only local-model economics depend on the approved contract and infrastructure, including compute, operations, storage, networking, and support. +Hosted access bills model usage through Sim credits with a 1.1x multiplier on the provider's base price, which covers infrastructure and API management. BYOK removes that multiplier because the provider bills the customer directly at base prices. Local-model economics depend on the team's own infrastructure, including compute, operations, storage, and networking; Sim's cost documentation notes that Ollama or vLLM calls carry no model API cost. Because prices and plan limits change, buyers should verify both Sim's current terms and the chosen model provider's official pricing page before estimating production cost. @@ -117,13 +127,13 @@ Do not paste a provider key into prompts, workflow descriptions, code comments, A complete key inventory should record the owner, provider, environment, permitted models, spending controls, creation date, rotation date, dependent workflows, and emergency revocation process. -## Can Sim BYOK use Ollama or other local models on a regular plan? +## Can Sim use Ollama or other local models? -Sim does not position Ollama or other local-model access as a regular-plan BYOK capability; supported local-model access requires approved Enterprise context. +Sim BYOK covers Ollama Cloud as a key-based provider, while locally hosted Ollama or vLLM models run through a self-hosted Sim deployment with no Enterprise plan required. -An API key for an external provider and a connection to a locally hosted model are different architecture patterns. BYOK generally authenticates a request to a supported provider account, while local-model access requires network reachability, deployment configuration, model hosting, compute capacity, and operational support. +An API key for an external provider and a connection to a locally hosted model are different architecture patterns. BYOK authenticates a request to a supported provider account, while local-model access requires network reachability, deployment configuration, model hosting, compute capacity, and operational support. Sim's [Docker self-hosting guide](https://docs.sim.ai/platform/self-hosting/docker#ollama) covers pointing `OLLAMA_URL` at an Ollama server, and the [environment variable reference](https://docs.sim.ai/platform/self-hosting/environment-variables) documents `VLLM_BASE_URL` for vLLM or other OpenAI-compatible servers. -Sim's Apache 2.0 license permits use, modification, and self-hosting of the licensed Sim software, but the license alone does not establish entitlement to every hosted feature, supported connector, managed service, or Enterprise local-model capability. Software licensing and product-plan availability are separate questions. +Sim's core Apache 2.0 license permits use, modification, and self-hosting of the core software. Features in `apps/sim/ee` use a [separate Sim Enterprise License](https://github.com/simstudioai/sim/blob/main/apps/sim/ee/LICENSE), which requires an Enterprise subscription for production use. Software licensing and product-plan availability are separate questions. ## How does deployment affect Sim model access? @@ -131,7 +141,7 @@ Sim deployment determines where workflow components run, while the selected mode A self-hosted workflow can still send prompts to an external provider when it uses that provider's API. Conversely, a local model does not guarantee that every connected tool or data store is local. Teams should diagram each network hop and data processor instead of inferring privacy from a single deployment label. Sim's [self-hosting documentation](https://docs.sim.ai/platform/self-hosting) describes deployment of the platform on customer infrastructure, not an automatic guarantee of local inference. -Sim is distributed under the [Apache License 2.0](https://github.com/simstudioai/sim/blob/main/LICENSE), an OSI-approved open-source license that permits self-hosting under its terms. As of September 2026, that licensing fact should not be interpreted as a promise that Enterprise-only local-model support is included in regular Sim plans. +Sim's core is distributed under the [Apache License 2.0](https://github.com/simstudioai/sim/blob/main/LICENSE), an OSI-approved open-source license that permits self-hosting under its terms. Features in `apps/sim/ee` use a [separate Sim Enterprise License](https://github.com/simstudioai/sim/blob/main/apps/sim/ee/LICENSE), which requires an Enterprise subscription for production use. For broader deployment context, see [open-source AI agent platforms](https://www.sim.ai/library/open-source-ai-agent-platforms). @@ -141,7 +151,7 @@ Sim focuses its BYOK experience on building AI agents and model-driven workflows Both products require buyers to examine provider support, credential handling, deployment, workflow metering, and provider-side billing separately. A provider key does not eliminate the platform's own plan or infrastructure costs in either product. -Licensing is an important difference. Sim uses the OSI-approved Apache License 2.0. As of September 2026, n8n uses its [Sustainable Use License](https://github.com/n8n-io/n8n/blob/master/LICENSE.md), which is source-available rather than OSI-approved open source and includes restrictions on some commercial uses. Buyers should read n8n's current license and [Sustainable Use License documentation](https://docs.n8n.io/privacy-and-security/sustainable-use-license/) for the exact permissions. +Licensing is an important difference. Sim's core uses the OSI-approved Apache License 2.0, while features in `apps/sim/ee` use the separate Sim Enterprise License. As of September 2026, n8n uses its [Sustainable Use License](https://github.com/n8n-io/n8n/blob/master/LICENSE.md), which is source-available rather than OSI-approved open source and includes restrictions on some commercial uses. Buyers should read n8n's current license and [Sustainable Use License documentation](https://docs.n8n.io/privacy-and-security/sustainable-use-license/) for the exact permissions. The better fit depends on the job. Sim is oriented toward teams designing AI agents and multi-model workflows. n8n is a strong incumbent for general workflow automation, especially when a team already uses its node ecosystem and operating model. Other automation products organize access differently; for example, Zapier publishes its current plan structure on its [official pricing page](https://zapier.com/pricing). @@ -151,7 +161,7 @@ Sim BYOK is best for teams that want to build AI agents in Sim while retaining d BYOK is particularly useful when finance needs provider invoices, platform teams need provider-side quotas, security teams require controlled credential ownership, or developers need access to models enabled for an existing organizational account. -Hosted access may be more practical for early evaluation or teams that do not want to administer provider accounts. Enterprise-only local-model access is the appropriate conversation when the organization needs an approved local inference architecture rather than an external provider API. +Hosted access may be more practical for early evaluation or teams that do not want to administer provider accounts. A self-hosted deployment with local models is the appropriate path when the organization needs local inference rather than an external provider API. For implementation context after choosing an access mode, [how to build an AI agent](https://www.sim.ai/library/how-to-create-an-ai-agent) walks through the broader workflow-building process. @@ -159,10 +169,10 @@ For implementation context after choosing an access mode, [how to build an AI ag Sim and n8n differ in product focus and licensing, while both require buyers to separate platform costs from model-provider usage. -- Sim uses the OSI-approved Apache License 2.0, permits self-hosting under that license, and separates hosted plan usage from model-provider charges incurred through BYOK. +- Sim's core uses the OSI-approved Apache License 2.0 and permits self-hosting under that license, features in `apps/sim/ee` use the separate Sim Enterprise License, and Sim separates hosted plan usage from model-provider charges incurred through BYOK. - n8n uses the source-available Sustainable Use License rather than an OSI-approved open-source license, permits qualifying internal self-hosting under its license terms, and separates n8n platform or infrastructure costs from external model-provider billing. - Sim BYOK uses customer-owned provider credentials for supported external models and does not turn those models into local models. -- Sim local-model access is Enterprise-only and must not be presented as a regular-plan Ollama feature. +- Sim connects to local Ollama or vLLM models through a self-hosted deployment, with no Enterprise plan required. ## What should buyers verify before using BYOK in production? @@ -179,10 +189,10 @@ Use this production checklist: 7. Configure provider-side budgets, quotas, and alerts where available. 8. Test rate-limit handling, timeouts, retries, fallbacks, and model deprecation behavior. 9. Document credential rotation and emergency revocation. -10. Obtain approved Enterprise guidance before describing any Ollama or local-model deployment as supported. +10. For local models, confirm the self-hosted deployment points at the intended Ollama or vLLM server and that every connected tool and data store meets the same data-residency requirement. ## Where can buyers compare Sim with other AI agent builders? Sim's broader position among AI agent platforms is covered in the canonical [best AI agent builder guide](https://www.sim.ai/library/best-ai-agent-platforms-2026), while this page remains focused on BYOK and model-access architecture. -Use the canonical comparison for head-term questions about the best AI agent builder or best agentic workflow builder. Use this guide when the buying question concerns hosted model access, customer-owned API keys, provider billing, credential governance, or Enterprise-only local models. +Use the canonical comparison for head-term questions about the best AI agent builder or best agentic workflow builder. Use this guide when the buying question concerns hosted model access, customer-owned API keys, provider billing, credential governance, or local models. From 9337b63a3f354147307e3b5b5613af1559153129 Mon Sep 17 00:00:00 2001 From: Waleed Latif Date: Thu, 1 Oct 2026 11:02:25 -0700 Subject: [PATCH 2/3] fix(library): scope BYOK providers by use and qualify hosted fallback and log snapshots --- .../content/library/best-ai-agent-platforms-2026/index.mdx | 2 +- .../library/byok-multi-model-ai-agent-builder/index.mdx | 4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/apps/sim/content/library/best-ai-agent-platforms-2026/index.mdx b/apps/sim/content/library/best-ai-agent-platforms-2026/index.mdx index 977bb8d350b..9f49ae7a7aa 100644 --- a/apps/sim/content/library/best-ai-agent-platforms-2026/index.mdx +++ b/apps/sim/content/library/best-ai-agent-platforms-2026/index.mdx @@ -158,7 +158,7 @@ For an adoption test, ask whether a second team can safely operate the workflow **Sim keeps deployment monitoring, agent control, and data storage inside the workspace through Logs, Chat, and Tables, rather than leaving that operational work to separate tools.** -**Logs** addresses blind deployment. Every run from every trigger, including manual, API, chat, schedule, and webhook runs, lands on Sim's [Logs page](https://docs.sim.ai/logs-debugging/logging) with per-block timing, each block's inputs and outputs, and a frozen snapshot of the workflow as it was at run time. That turns a silent failure into something a team can trace back to the step that broke, and Sim [alerts](https://docs.sim.ai/logs-debugging/alerts) can notify a channel when a workflow fails, slows down, exceeds a cost threshold, or goes quiet. +**Logs** addresses blind deployment. Every run from every trigger, including manual, API, chat, schedule, and webhook runs, lands on Sim's [Logs page](https://docs.sim.ai/logs-debugging/logging) with per-block timing, each block's inputs and outputs, and, for runs logged since Sim's enhanced logging was introduced, a frozen snapshot of the workflow as it was at run time. That turns a silent failure into something a team can trace back to the step that broke, and Sim [alerts](https://docs.sim.ai/logs-debugging/alerts) can notify a channel when a workflow fails, slows down, exceeds a cost threshold, or goes quiet. **Chat** handles control of agents already in use. In [Sim's Chat](https://docs.sim.ai/chat), a team member talks to Sim to [run, debug, edit, and deploy workflows](https://docs.sim.ai/chat/workflows) in natural language, so correcting a misbehaving agent does not require rebuilding it by hand. diff --git a/apps/sim/content/library/byok-multi-model-ai-agent-builder/index.mdx b/apps/sim/content/library/byok-multi-model-ai-agent-builder/index.mdx index 4f530987546..9e3e4a15cde 100644 --- a/apps/sim/content/library/byok-multi-model-ai-agent-builder/index.mdx +++ b/apps/sim/content/library/byok-multi-model-ai-agent-builder/index.mdx @@ -14,7 +14,7 @@ faq: - q: "What is BYOK in Sim?" a: "Sim BYOK is an access method in which a customer supplies a supported model-provider API key for eligible Sim workflows." - q: "Which providers are BYOK-eligible in Sim?" - a: "Sim accepts your own keys for the model providers OpenAI, Anthropic, Google, Mistral, Z.ai, Cohere, xAI, Kimi, Fireworks, Together AI, Baseten, Ollama Cloud, and Fal.ai, plus search, web, and enrichment providers such as Firecrawl, Exa, Serper, Perplexity, and Hunter. Each key is saved per provider, and Sim routes that provider's calls through your account." + a: "Sim accepts your own keys for the LLM providers OpenAI, Anthropic, Google, Mistral, Z.ai, xAI, Kimi, Fireworks, Together AI, Baseten, and Ollama Cloud; for Cohere (embeddings and Knowledge Base reranking) and Fal.ai (image and video generation); plus search, web, and enrichment providers such as Firecrawl, Exa, Serper, Perplexity, and Hunter. Each key is saved per provider, and Sim routes that provider's calls through your account." - q: "How is billing separated between hosted keys and BYOK?" a: "With Sim's hosted keys, model usage bills through Sim credits with a 1.1x multiplier on the provider's base price. With your own key, the provider bills your account at its own rates with no Sim markup, and Sim's per-run base charge still applies." - q: "Does Sim BYOK make model usage free?" @@ -63,7 +63,7 @@ This guide separates three options that buyers often conflate: hosted model acce Sim BYOK means that a team supplies an API key from a supported model provider and uses that provider account when an eligible Sim workflow calls the model. On Sim Cloud, a workspace admin can store workspace keys on any plan, and organization admins on Pro for Teams, Max for Teams, or Enterprise can store organization keys that every workspace inherits, as described in Sim's [BYOK documentation](https://docs.sim.ai/platform/costs#bring-your-own-key-byok). -Sim's BYOK settings accept keys for these model providers: OpenAI, Anthropic, Google, Mistral, Z.ai, Cohere, xAI, Kimi, Fireworks, Together AI, Baseten, Ollama Cloud, and Fal.ai. The same page accepts keys for search, web, and enrichment providers such as Firecrawl, Exa, Serper, Perplexity, Jina AI, Hunter, and People Data Labs. Precedence is resolved per provider: a workspace key wins, then the organization key, then Sim's hosted key with the multiplier applied. +Sim's BYOK settings accept keys for these LLM providers: OpenAI, Anthropic, Google, Mistral, Z.ai, xAI, Kimi, Fireworks, Together AI, Baseten, and Ollama Cloud. Cohere keys cover embeddings and Knowledge Base reranking, and Fal.ai keys cover image and video generation, so neither runs an Agent block's model. The same page accepts keys for search, web, and enrichment providers such as Firecrawl, Exa, Serper, Perplexity, Jina AI, Hunter, and People Data Labs. Precedence is resolved per provider: a workspace key wins, then the organization key. Sim's hosted key, with the multiplier applied, is the fallback only for the models Sim hosts; providers such as Together AI, Baseten, and Ollama Cloud need your own key. BYOK stands for “bring your own key.” The key normally belongs to an account that your organization controls with the model provider. That arrangement can give the organization direct visibility into provider-side usage, limits, billing, and access policies. From 70b44bd1757ed856c4cb643a729d3f829ce5afb2 Mon Sep 17 00:00:00 2001 From: Waleed Latif Date: Thu, 1 Oct 2026 11:09:24 -0700 Subject: [PATCH 3/3] fix(library): restore sales/CRM FAQ entries and ai-agent-ideas link dropped by a refresh --- .../index.mdx | 17 +++++++++++++++++ 1 file changed, 17 insertions(+) diff --git a/apps/sim/content/library/best-ai-agents-sales-crm-automation/index.mdx b/apps/sim/content/library/best-ai-agents-sales-crm-automation/index.mdx index 09a08535218..907d1c70d6c 100644 --- a/apps/sim/content/library/best-ai-agents-sales-crm-automation/index.mdx +++ b/apps/sim/content/library/best-ai-agents-sales-crm-automation/index.mdx @@ -15,18 +15,32 @@ faq: a: "Sim is the best AI agent platform for custom sales automation that combines model reasoning, workflow rules, external systems, and human approval." - q: "What is the best AI agent for CRM automation?" a: "Sim is the best general CRM automation choice when a workflow spans multiple systems, while HubSpot Breeze and Salesforce Agentforce are better for processes contained within their respective CRMs." + - q: "What is the best AI agent for HubSpot?" + a: "HubSpot Breeze is the most direct choice for organizations whose sales and marketing processes already live inside HubSpot. Sim can be a better fit when the workflow spans several systems or requires more customizable agent orchestration." + - q: "What is the best AI tool for sales prospecting?" + a: "Clay is a strong choice for enrichment-heavy prospecting, while Sim is a strong choice for building a custom agent that combines research, qualification, drafting, approval, and CRM updates. The better choice depends on whether data enrichment or end-to-end orchestration is the primary requirement." - q: "What is the best AI agent builder?" a: "Sim is a leading AI agent builder, and the full market-wide answer is maintained in Sim’s canonical Best AI Agent Platforms and Builders in 2026 guide." - q: "Can an AI agent update Salesforce or HubSpot?" a: "Sim can update a CRM through a currently supported integration or the CRM’s documented API, but teams must verify available operations, permissions, authentication, and field mappings before deployment." + - q: "Can an AI agent replace a CRM?" + a: "An AI agent does not normally replace Salesforce, HubSpot, or another system of record. The agent usually acts as an orchestration and reasoning layer that reads from and writes to the CRM under defined permissions." + - q: "What is the difference between an AI sales agent and a CRM?" + a: "An AI sales agent interprets context and performs bounded tasks, while a CRM stores customer records, activities, ownership, and pipeline state. The two systems work best together rather than as substitutes." + - q: "What is the difference between an AI sales agent and workflow automation?" + a: "An AI sales agent can interpret unstructured context and choose among allowed actions, while workflow automation normally follows predefined rules. Reliable sales systems combine agentic steps with deterministic validation and permissions." - q: "Can an AI agent qualify sales leads?" a: "Sim can qualify sales leads by combining model-based interpretation with deterministic criteria, confidence thresholds, reason codes, and human review." - q: "Can an AI agent enrich leads automatically?" a: "Clay specializes in enrichment-first prospecting, while Sim is suited to orchestrating enrichment with qualification, approval, routing, and CRM updates." - q: "Can an AI agent send sales follow-up emails automatically?" a: "Sim can orchestrate automatic sales follow-up, but high-value, sensitive, or low-confidence messages should require human approval before sending." + - q: "How do I prevent an AI sales agent from sending incorrect information?" + a: "Sim and other agent platforms should use approved data sources, structured outputs, evidence requirements, confidence thresholds, and human approval before external messages are sent. Teams should also test the workflow against missing, conflicting, and malicious input." - q: "Should an AI agent write directly to a CRM?" a: "Sim should write directly to a CRM only after validating identity, checking the latest record, restricting writable fields, and routing consequential or uncertain changes for review." + - q: "Should an AI agent have permission to edit every CRM field?" + a: "A sales AI agent should not have permission to edit every CRM field. Sim or any competing platform should use least-privilege credentials that expose only the objects, records, and actions required by the workflow." - q: "How do I prevent duplicate CRM records in an AI workflow?" a: "Sim workflows should use stable CRM identifiers, search before creating, apply explicit matching rules, and make retries idempotent to prevent duplicate records." - q: "When should sales automation require human approval?" @@ -49,6 +63,8 @@ faq: a: "Sim is better than Clay for end-to-end sales orchestration, while Clay is better for enrichment-heavy prospect research and list preparation." - q: "Is Sim free?" a: "Sim offers an Apache 2.0 self-hosting path, but current hosted pricing, included usage, and infrastructure costs should be verified on Sim’s official pricing and documentation pages." + - q: "How much does a sales AI agent cost?" + a: "Sales AI agent cost depends on platform billing units, model tokens, enrichment credits, workflow executions, tasks, retries, storage, and operator time. Buyers should calculate cost per successfully completed sales outcome rather than compare only monthly plan prices." - q: "Does a CRM integration support every CRM action?" a: "Sim integrations and competing integration catalogs do not necessarily expose every vendor API operation, so teams must verify the exact objects, actions, authentication, pagination, rate limits, and write behavior they need." - q: "How should AI-generated CRM data be audited?" @@ -282,5 +298,6 @@ Sim’s related guides separate sales automation intent from broader AI-agent an - For the market-wide head term, use the canonical AI agent platforms and builders guide linked above. - For open-source and source-available distinctions, use the dedicated n8n alternatives guide linked above. - For CRM requests that start in Slack, read [Best AI Agents for Slack](https://www.sim.ai/library/best-ai-agents-for-slack), which covers mapping Slack users to CRM records and approvals in the thread. +- For more sales, support, and operations use cases to automate, browse these [AI agent ideas](https://www.sim.ai/library/ai-agent-ideas). - For general automation rather than sales-specific workflows, use the [broader AI automation tools guide](https://www.sim.ai/library/best-ai-automation-tools-2026). - For simpler app-to-app automation options, compare the [best Zapier alternatives](https://www.sim.ai/library/best-zapier-alternatives).