SmythOS
smythos.com
- Free plan
- Open source
- Windows
- macOS
- Linux
- Web
- iPhone and iPad
- Android
- API
- Self-hosted
What SmythOS is
SmythOS is an intelligence orchestration platform for designing, observing, and deploying AI agents. Its Studio uses a drag-and-drop canvas to build agents and workflows without code, while Agent Weaver can create, troubleshoot, and update workflows through conversational prompts. Agents can be delivered through chat, APIs, webhooks, MCP servers, or ChatGPT, and can also be exported to run locally. Integrations include Slack, Trello, GitHub, and Stripe, as well as OpenAPI and Postman collections. SmythOS supports models from OpenAI, Anthropic, Hugging Face, and customer-provided sources. Deployment options include SaaS, local, on-premises, and VPC; its security materials describe least-privilege controls and isolated agent contexts. Built-in model usage is billed separately according to usage. The Public plan is free and includes one seat; Private Builder costs $39.00 per seat per month and includes 100 fast API calls per day. Other listed plans have higher quotas and features. SmythOS supports web, Windows, macOS, Linux, API, and self-hosted environments, and is headquartered in Houston, Texas.
Who it suits
It suits people and teams building AI agents, from visual workflow authors to developers integrating agents through APIs or running them locally. Teams choosing a paid tier should account for API-call limits and separate model usage charges.
Good to know, and what to watch
- Visual agent and workflow builder requires no code.
- Agent Weaver works from conversational prompts.
- Supports local, on-premises, and VPC deployment.
- Integrations include Slack, GitHub, Trello, and Stripe.
- Open-source runtime and SDK has more than 40 components.
- Built-in model usage is billed separately.
- Builder includes 100 fast API calls per day.
- Private Builder costs $39.00 per seat per month.
Plans
Every plan and pricePublic Free
Builder
Startup
Scaleup
Free plan
Getting started
- Open https://smythos.com/
- Choose a free or paid plan
- Use Studio's drag-and-drop canvas to build an agent or workflow
- Create or revise workflows with Agent Weaver prompts
- Connect supported services or model providers
- Deploy through chat, APIs, webhooks, MCP servers, ChatGPT, or local export
Limits to check
The free Public plan has one seat and includes $5 in monthly free credits, with model usage charged at 2.5 times the model cost. Builder provides 100 fast API calls per day; other quotas, seats, and features vary by plan, while model usage remains separately billed according to usage.
The Club's review
SmythOS is an intelligence-orchestration platform for designing, observing and deploying AI agents. It suits teams that need visual and conversational agent building alongside flexible hosting and integrations. The trade-off is a usage model in which API allowances and model costs sit alongside the subscription.
Overview
SmythOS combines a drag-and-drop Studio with Agent Weaver, a prompt-driven tool that can create, troubleshoot and update workflows. That gives non-programmers a visual route and lets experienced users describe revisions in ordinary language. Agents can move from experimentation to chat, API, webhook, MCP-server or ChatGPT delivery, or be exported for local operation. The breadth is useful for organizations with varied deployment requirements, but less suitable for buyers wanting one predictable, all-inclusive bill.
Scheduled runs and web browsing extend the platform beyond manual chat interactions. Its open-source Runtime Environment and SDK, described as containing more than 40 components, provide a production-oriented path for teams that want runtime control rather than a purely hosted service.
Key features
Visual and conversational building
Studio’s canvas lets users assemble agents and workflows without writing code. Agent Weaver provides a second workflow: explain what should be built or changed, then use conversational troubleshooting and updates. These approaches should appeal to mixed teams, although highly specialized behavior may still require technical oversight.
Integrations and model choice
Prebuilt connectors include Slack, Trello, GitHub and Stripe. OpenAPI and Postman collection support gives teams a way to connect additional services without waiting for a dedicated connector. SmythOS supports providers including OpenAI and Anthropic, Hugging Face models and customer-provided models, so organizations can align agents with existing model strategies. Built-in model usage is charged separately, however, making frequent or model-heavy workloads more expensive than the subscription alone suggests.
Deployment, security and data control
Deployment options span hosted channels and exported local operation, with local, on-premises and VPC modes for customers that need data to remain in their own environment. Security controls include zero-trust access, least privilege, fine-grained ACLs and isolated agent contexts. SaaS traffic uses TLS 1.2 or newer, while stored data is encrypted through the cloud provider’s KMS. HashiCorp Vault and AWS Secrets Manager integrations address secret handling. These controls are relevant to regulated or security-conscious teams, while smaller users may not need the operational complexity of self-hosting.
Pricing
SmythOS uses freemium pricing, with paid plans from $39 per user per month. Built-in model usage is billed separately according to consumption, so API quotas and model discounts should be evaluated together.
| Plan | Price and allowances | Best fit |
|---|---|---|
| Public Free | Free, billed monthly; 1 seat | Individual evaluation with a single seat. |
| Builder | $39 a month, billed monthly; 100 fast API calls per day | One builder with modest daily automation needs. |
| Startup | $399 a month, billed per month; $200 in free credits per month, 50% model-usage discount, 5,000 fast API calls per day, 10 team spaces, 3 seats, RAG Agents and collaboration features | Small teams moving from experiments to shared agent work. |
| Scaleup | Starting at $1,499 a month; $300 in free credits per month, 60% model-usage discount, 25,000 fast API calls per day, 50 team spaces, 5 seats, forward-deployed engineer, priority support and white labeling | Growing deployments needing support and branded delivery. |
| Enterprise | Starting from $4,955 a month; 100% model-usage discount, unlimited agents, tasks, fast API calls, team spaces and seats, on-premises/VPC deployment, OEM distribution, compliance tailoring and custom retention policies | Large organizations requiring scale, deployment control and tailored governance. |
A separate Public plan costs nothing and is totally free, with $5 in monthly credits, one seat, public agents, the visual IDE, bring-your-own-model support and free local deployment; model usage costs 2.5 times the standard rate. Private Builder costs $39 a month per seat and includes $20 in monthly credits, a 40% model-usage discount, 100 fast API calls per day, private agents and unlimited Weaver AI messages. The free tiers are useful for trying the product, but their single-seat structure, limited credits or higher model rate constrain sustained production use.
Platforms
SmythOS runs through the web and supports API, Linux, macOS, Windows and self-hosted deployments. That combination suits teams spanning browser users, local environments and controlled infrastructure rather than a single-device workflow.
Who it's for
Choose SmythOS if your organization needs several agent-building methods, broad deployment routes, integrations and the option to keep data in a local, on-premises or VPC environment. Startup is the practical entry point for a small collaborative team; Scaleup and Enterprise target organizations that need higher quotas, priority or forward-deployed support, white labeling, OEM distribution or customized compliance. A solo user with light usage should start with Public Free or Builder, while anyone unwilling to track model charges should look elsewhere.
Pros and cons
- Pro: Visual Studio and conversational Agent Weaver cover both no-code assembly and prompt-led workflow changes.
- Pro: Chat, API, webhook, MCP, ChatGPT and local deployment options accommodate different delivery architectures.
- Pro: On-premises/VPC support, isolated contexts and vault integrations address demanding security environments.
- Con: Fast API quotas range from 100 per day to unlimited, so lower plans can restrict active automations.
- Con: Built-in model usage is charged separately; the Public plan also applies a 2.5x model-usage cost.
- Con: Entry plans provide only one seat, while meaningful collaboration begins at the $399 Startup tier.
Alternatives
Houston is a better fit for a personal workspace of up to three people, limited usage and access to more than 1,000 integrations on a free plan. Pick QuickBlox Chat API when the priority is embeddable chat across Android, iOS, web or self-hosted environments, with a free tier for 500 users. Lindy suits users who prefer a web or browser-extension product priced by monthly credits per user. Microsoft Copilot is the more natural choice for a general assistant across Microsoft-supported desktop and mobile platforms. Automation Anywhere Document Automation is aimed at document workflows, including a community tier for eligible small businesses. Relevance AI offers a web-only free workforce with daily credits and action limits. UiPath Test Cloud is the alternative for testing-focused teams needing self-hosted and cloud options. Agentforce Vibes fits organizations already working in Salesforce-style credit or flat-fee arrangements.
For broader comparisons, see the AI Agent Builders and AI Agent Platforms directories.
Verdict
SmythOS is a strong choice for teams that want visual and prompt-based agent creation with integrations, multiple deployment paths and enterprise-grade control options. Choose it when deployment flexibility and governance matter more than a simple bill. Look elsewhere if your usage is light, you need collaboration below three seats, or separate model charges and fast-call quotas make budgeting harder.
The verdict
SmythOS offers visual and prompt-driven agent creation, several deployment routes, and integrations for connecting workflows. Review the API-call allowance and model charges alongside your deployment and team needs when selecting a plan.
Questions people ask
Is SmythOS free to use?
Yes. Public Free and Public plans are listed as free, and paid plans start at $39 a month.
What platforms does SmythOS support?
It supports web, Windows, macOS, Linux, API, and self-hosted environments.
Can SmythOS run inside a private environment?
Yes. The product supports local, on-premises, and VPC deployment, with data able to remain in the customer's environment.
Which models can SmythOS use?
It lists models from OpenAI, Anthropic, Hugging Face, and customer-provided models.
Does SmythOS have an open-source component?
Its Runtime Environment is described as an open-source runtime and SDK for production AI agents.
What the maker publishes
- Agent architecture
- Agents combine components, skills, logic, memory, orchestration, prompts, and access controls and can operate through API, chat, voice, batch, or embedded interfaces. source
- Agent creation
- Agent Weaver can create, troubleshoot, and update agent workflows through conversational prompts. source
- Agent Weaver
- The product documentation describes Weaver as a prompt-driven way to create agents and workflows without coding. source
- API surfaces
- Published agents expose REST and WebSocket APIs, and SmythOS supports API, webhook, OpenAI-compatible LLM, and MCP deployments. source
- CLI
- The SRE CLI is installed with npm using @smythos/cli and scaffolds new SDK projects with the sre create command. source
- Company headquarters and size
- LinkedIn lists SmythOS as a privately held Houston, Texas technology company with 11–50 employees. source
- Data protection
- For SmythOS SaaS, the security page states that data in transit uses TLS 1.2+ and data at rest is encrypted using the cloud provider’s KMS. source
- Deployment
- Agents can be deployed through chat, APIs, webhooks, MCP servers, and ChatGPT, or exported to run locally. source
- Deployment isolation
- The product supports cloud, on-premises or VPC, and on-device deployments, including local-model operation for keeping data inside customer infrastructure. source
- Documentation
- The documentation site covers Studio, Weaver, Runtime, deployments, collaboration, templates, account management, SDK usage, components, integrations, and self-hosting. source
- Encryption
- The security documentation states that SaaS traffic uses TLS 1.2 or newer and stored data uses cloud-provider key-management encryption. source
- Founded year
- Built In lists SmythOS as founded in 2023 with 50 total employees. source
- Headquarters
- Houston, Texas, United States source
- Integrations
- The integrations page lists prebuilt connections such as Slack, Trello, GitHub, and Stripe, alongside support for OpenAPI and Postman collections. source
- Integrations
- Documented integrations include OpenRouter, Replicate, Stability AI, ElevenLabs, PlayHT, Perplexity, Tavily, Fal.ai, Hugging Face, Google Translate, and OpenAPI. source
- Legal identity and founders
- Crunchbase lists the legal name as SmythOS, a sub-organization of INK Content Inc., founded by Alexander De Ridder, Gary Haymann, and Michael Umansky. source
- License
- The SRE runtime, SDK, and CLI repository is open source under the MIT License. source
- Limit: interface complexity
- A G2 reviewer says the interface can become cluttered on complex workflows and asks for a lower-noise advanced mode. source
- Limit: refunds
- A Trustpilot reviewer complained that there was no direct refund option; the company replied that paid-plan refunds are handled case by case. source
- Model charges
- The pricing page says built-in model usage is billed separately from the subscription according to usage. source
- Models
- SmythOS says it supports models from providers including OpenAI and Anthropic, plus Hugging Face models and customer-provided models. source
- Notable news: open sourcing
- A June 17, 2025 press release announced SmythOS's transition to open source, beginning with the SRE runtime, SDK, CLI, and new documentation. source
- OpenAI-compatible API
- LLM deployment provides versioned OpenAI-compatible chat-completions endpoints callable from Python, Node.js, Make, Zapier, n8n, or any OpenAI SDK. source
- Open source runtime
- SmythOS describes its Runtime Environment as an open-source runtime and SDK for production AI agents, with more than 40 components. source
Checked against the maker's pages on 14 Oct 2026.