Ekinox is a visual AI automation platform that arrived in late 2025. It aims to let anyone build intelligent workflows without writing code. You drop blocks on a canvas and connect them into agents that can search the web update spreadsheets send emails or call an API.
It supports dozens of AI models from major providers and integrates with popular business apps.
Its plans range from a free tier for small experiments to an enterprise package with on‑premise deployment. In this ekinox review, you will learn what Ekinox does, how much it costs, and whether it fits your needs.
What Ekinox is and who it’s for
Ekinox markets itself as an AI workflow builder. Unlike generic automation tools that added AI later it was designed around agents from the start.
Each workflow is a graph of tasks that can reason on data call APIs make decisions or ask humans for approval. Because the interface is visual and block based you do not write code.
That makes it accessible to marketers support teams analysts and anyone curious about AI. At the same time it offers deep customisation and advanced controls that appeal to developers and enterprise architects.
Key features
Visual canvas
The heart of Ekinox is its drag and drop canvas. You choose blocks for agents APIs loops conditions and routing then join them with lines. This view lets you see logic at a glance. You can zoom out for a high level flow or click into a block to inspect inputs and outputs. The canvas supports parallel branches so complex paths run side by side. If you need to reuse a flow you can wrap it as a sub‑workflow block and drop it into other canvases.
AI model flexibility
Ekinox connects to multiple AI providers. You can pick GPT‑4 or GPT‑4o from OpenAI for language tasks and mix in Claude from Anthropic or Gemini from Google for other jobs. The platform also supports Mistral Groq Hugging Face models and local models via Ollama. For speech there are integrations with ElevenLabs. Because you select a model per task you can optimise quality cost or speed. Advanced users may bring their own API keys to avoid markups.
Built‑in integrations
There are connectors for over one hundred services. Communication tools include Gmail Slack Outlook Discord Telegram WhatsApp and Microsoft Teams. Productivity integrations cover Notion Google Sheets Airtable Jira Linear Monday.com and more. You can pull data from databases like PostgreSQL MySQL Supabase Pinecone and Qdrant and then send outputs back. Web search connectors allow real time retrieval through Google Search Perplexity or Firecrawl. If you need a custom source you can use the platform’s Model Context Protocol to build your own.
Triggers and deployment
Workflows can start from many triggers. You can run them manually from the editor test them via a chat interface or expose them as REST endpoints for integration. Scheduled jobs allow recurring runs on a daily or hourly basis. Webhooks let external events such as a form submission or GitHub push kick off a flow. A workflow can also be embedded in a chat bot that watches Slack or Teams channels.
AI Copilot
Inside the editor you have an assistant called Copilot. It analyses your canvas suggests blocks explains what each part does and can even edit flows with your approval. You can type a short prompt like “add summarisation after scraping” and Copilot will insert the right tool. Copilot uses tokens from your workspace credits so its usage contributes to your plan’s limits. It feels like having an AI pair programmer for automation.
Knowledge base and RAG
For retrieval augmented generation Ekinox comes with a knowledge base. You can upload documents in formats like PDF Word or Markdown and it will create vector embeddings using pgvector. Agents can then search the knowledge to answer questions cite sources or enrich data. You can tweak chunk size edit or tag entries and filter by metadata. This makes it possible to build chatbots or support tools that use your own data.
Collaboration and permissions
Teams can edit the same workflow at the same time. The canvas shows who is viewing or typing on each block. Comments and version history help track changes. Roles determine what each person can do. Read‑only users can view flows and logs but cannot run or edit. Write users can build and deploy workflows but not manage settings. Admin users can invite or remove others adjust environment variables and alter workspace details. Enterprise plans add SSO and SAML integration for single sign on.
Logging and monitoring
Ekinox collects detailed logs for every run. A real time console in the editor shows each block as it executes with success or error status and timing.
A separate logs page lists all runs across all triggers with filters for date status or workflow. Clicking a log reveals input and output at each block formatted as JSON or Markdown.
You can also view the execution timeline and inspect variable values. Log retention depends on the plan. Free users get seven days of history Pro users get thirty days Team users get ninety days and Enterprise users have unlimited retention.
Cost transparency
Each execution has a base charge of one tenth of a cent. If you use AI models there is an additional cost per token.
The platform displays the token usage of each model and calculates the cost by multiplying tokens by the provider’s price.
When you use Ekinox’s hosted models there is a multiplier (currently 2.5×) to cover infrastructure. You can bring your own keys to pay standard rates. Workflows that only call APIs or run logic blocks incur just the base charge.
Pricing plans
Ekinox splits its offerings into four tiers.
- Free plan – This entry level tier costs nothing and includes five dollars of AI credits. You can run ten workflows per minute and access basic integrations like email and messaging. Community support is available via forum. Log retention is seven days. This plan is ideal for individuals who want to explore agent based automation without commitment.
- Pro plan – At thirty dollars per month you get thirty dollars of AI credits and unlimited workflows. All integrations are unlocked. Copilot becomes available and you receive priority support. Logs are kept for thirty days and advanced analytics reveal token use and cost patterns. This tier suits professionals or small teams that build production flows.
- Team plan – This tier costs one hundred dollars per month and includes one hundred dollars of AI credits. It adds real time collaboration features and team management. You get single sign on and SAML for identity management and log retention extends to ninety days. Ten workspaces are included so separate departments or projects can have their own areas. This plan targets small to medium businesses.
- Enterprise plan – At five hundred dollars per month you receive five hundred dollars of AI credits. Workspaces are unlimited. Dedicated 24 7 support and a custom service level agreement are part of the package. On‑premise deployment allows you to host Ekinox within your own infrastructure which may be vital for regulated industries. You can also request custom integrations and unlimited log retention. This plan is for large organisations that demand full control and support.
All plans include SSL encryption automatic backups and a 99.9 percent uptime guarantee. You can cancel at any time. The free plan does not require a credit card.
Pros
- Accessible and visual – The drag and drop interface lowers the barrier for non‑coders. You see your logic as a flowchart which reduces cognitive load. Templates help you start quickly.
- Multi‑model support – Few automation tools let you mix GPT Claude Gemini and other models in one flow. This flexibility lets you choose the right engine for each task and avoid vendor lock in.
- Rich integrations – With over one hundred connectors you can automate across emails chats databases spreadsheets CRM systems and more. The Model Context Protocol allows custom connections.
- Built‑in RAG – The native knowledge base for retrieval augmented generation enables internal chatbots without relying on external search. It stores your documents and indexes them automatically.
- AI Copilot – Having an assistant that understands your workflow and suggests improvements speeds up development. It can explain your flows in natural language which is useful for onboarding.
- Collaborative by design – Real time editing and granular permissions make it easy for teams to build together. You can add approval steps for human review.
- Transparent costs – The platform shows token usage and cost per execution. You can set budgets and alerts to avoid surprises. Bringing your own API keys reduces markups.
- Security and compliance – SOC2 alignment bank grade encryption and options for on‑premise deployment address privacy concerns. Logs redact API keys by default.
- Scalable plans – The free tier lets you test with no risk and paid tiers scale up with more credits workflows and features. Upgrading is straightforward.
Cons
- New and unproven – Ekinox only launched publicly in late 2025. There are few independent reviews or case studies. Early adopters may encounter bugs or missing features that mature platforms have solved.
- Credit based model – Each plan includes AI credits but heavy users will burn through them quickly. Once credits run out you must buy more or your workflows pause. Hosted models are priced at a premium because of the 2.5x multiplier.
- Limited free tier – The free plan restricts to ten workflows per minute and basic integrations. Log history is short. Serious projects may find it too constrained.
- Confusing integration counts – Marketing references to eighty or one hundred integrations change frequently. Some connectors may be in early access or require additional configuration.
- No external API for building flows – You must use the visual interface. Unlike tools such as Zapier or n8n there is no programmatic way to generate flows via an API. This may limit automation of the automation.
- Enterprise features locked behind cost – On‑premise deployment custom integrations and unlimited logs are exclusive to the enterprise plan. Smaller companies that require those features cannot access them without paying the highest price.
- Learning curve for complex flows – While basic flows are simple the platform introduces many block types triggers and settings. Users must understand AI reasoning loops concurrency and error handling to build robust agents.
Final verdict
Ekinox enters the market as a purpose built AI automation platform. Its visual canvas multi‑model support and rich integrations make it attractive for teams that want to build AI agents quickly.
The Copilot assistant and knowledge base set it apart from traditional workflow tools. Pricing is competitive when compared to hiring developers or building in house but the credit model requires careful monitoring. Because it is new there are uncertainties around reliability and support.
If you are a solo entrepreneur or small team curious about AI automation the free plan is a risk free way to experiment. For growing teams that need professional features the Pro or Team tiers provide value especially if you leverage multiple AI models and want an all in one environment.
Large enterprises should assess the security claims and integration roadmap but may find the on‑premise option compelling. Overall Ekinox shows promise as a next generation automation builder but it will need time and community feedback to mature.
