Viso is the computer vision company behind viso.ai. It offers two products built on the same visual general intelligence engine: Viso Now, a self-serve platform that turns a plain-language prompt into a live vision application, and Viso Suite, an enterprise platform that covers the full computer vision lifecycle from data and models to deployment and operations. Between them, the two products span everyday video analytics and end-to-end platform work, which is why the next question after “what is Viso?” is usually which of the two a team actually needs.
This guide explains both products, how agentic computer vision works inside each, and the signals that point a team toward one or the other. It is written for operations, safety, quality, and innovation leaders, as well as the IT and data teams who will be asked to support the platform they choose.
Viso, viso.ai, and Visual General Intelligence
Viso and viso.ai refer to the same company. Viso is the short name, and viso.ai is both its domain and its written brand name. The company is based in Switzerland and London, and its founders began building computer vision systems in 2014, long before they named Visual General Intelligence (VGI). Across years of customer deployments, they saw that the model was rarely the only bottleneck, because the higher cost came from rebuilding a computer vision system from scratch for every camera, environment, and use case.
VGI is the company’s answer to that problem. A narrow model can only find what it was trained to find, while a VGI engine reasons over a whole scene and can be pointed at a new question in plain language, a shift explained in VGI: from pixels to perception. viso.ai’s VGI-1 perception engine powers both of its products:
- Viso Now is a self-serve platform where anyone can describe an AI vision application in plain language and have it running within minutes.
- Viso Suite is an enterprise platform for building, deploying, and operating mission-critical computer vision across many sites, teams, and systems.

Bring a new AI vision application to life.
Video Analytics and a Full Lifecycle Platform
Industry comparisons often file viso.ai under video analytics, next to tools that package camera feeds into alerts and dashboards. That label is accurate, although it describes only part of what the two products cover together:
- Video analytics. Both products turn existing camera feeds and recorded footage into events, counts, alerts, and dashboards, the core of video analytics.
- The full computer vision lifecycle. Viso Suite also covers the stages that separate data, training, and deployment tools usually handle, including data collection, model training, application building, deployment, monitoring, and updates.
- Enterprise operations. Viso Suite connects cameras, sensors, edge devices, AI models, workflows, and business systems, so a detection can become a work order, a compliance record, or an ERP update.
The distinction matters during vendor evaluation. A team that needs one analytics outcome this month and an enterprise that needs a governed platform for dozens of vision applications are solving different problems, and viso.ai serves each with a product designed for it. Viso Suite also works with the AI models a company already runs, so earlier investments in data and training carry forward.
What Is Viso Now?
Viso Now is viso.ai’s self-serve platform for creating and running real-world AI vision applications without datasets, labeling, or model training. A user types what they want a camera to watch for, or drops in a video clip, and an agent builds the application around that description. The Viso Now product page describes the process in three steps:
- Start with an idea, written in plain language or shown with a short clip.
- Watch the agent connect the feed, recognize the scene, write the detection logic, and wire the alerts.
- Refine any rule, threshold, or output, then publish the application so it runs live.
Inputs include IP and CCTV cameras over RTSP and ONVIF, existing video systems and NVRs, MP4 files and webcams, and cloud folders on AWS, Azure, and GCP. Results can be pushed to Slack, Microsoft Teams, email, or a direct API, and dashboards track counts, compliance, trends, and deviations as new results arrive.
So what is Viso Now for? For decision makers, the practical answer is speed. A conventional computer vision project usually starts with collecting and labeling images and training a model before anything runs, while Viso Now starts from a written description, so a team can see whether an idea works on its own footage the same day it is proposed.
Pricing follows the same self-serve logic. There is a free plan with daily credits, paid plans add monthly credits that are shared across unlimited users, and the current tiers are listed on the Viso Now pricing page.

What Is Viso Suite?
Viso Suite is viso.ai’s all-in-one enterprise AI vision platform for building, deploying, and scaling mission-critical computer vision applications across an organization. It covers the full lifecycle of a vision application in one environment:
- Train: collect, manage, and refine data to build models tailored to the business.
- Build: compose and customize vision workflows in a visual no-code and low-code studio.
- Deploy: roll applications out across sites in the cloud, on premises, or at the edge, on hardware ranging from standard servers to NVIDIA Jetson edge devices.
- Operate: monitor performance in real time, manage updates, and keep security controls in place.
So what is Viso Suite for? It is built for the way production vision works in practice. The model is only one part of a working system, and most of the effort goes into everything around it, including connecting cameras, managing data, keeping applications running, and feeding results into the tools people already use. Viso Suite is built for that surrounding work, which is why it connects to existing cameras, sensors, and video systems and sends outputs into ERP platforms, BI tools, and other business systems.
Viso Suite is also delivered as a partnership. A dedicated viso.ai expert team works through discovery, a pilot deployment in the customer’s own environment, and scale, including a managed service once applications are live. Security is built into the platform, with SOC 2 Type II and ISO 27001 certification, SSO, role-based access control, isolated tenants, and a full audit trail, all detailed on the Viso Suite security page.
Viso Now vs Viso Suite: Key Differences
When teams compare Viso Suite vs Viso Now, the first point to settle is that they are separate products. The two share the engine and the agentic approach to building applications, then diverge on almost everything that surrounds the application once it is live. The table below summarizes the differences that tend to decide the choice.
| Dimension | Viso Now | Viso Suite |
|---|---|---|
| Category | Agentic video analytics and vision application builder | Enterprise AI vision platform for the full computer vision lifecycle |
| Primary builder | Operations, safety, quality, and innovation teams, from single sites to enterprise departments | Enterprise teams working with a dedicated viso.ai expert team |
| How applications are built | Plain-language prompt or video clip, assembled by an agent | Visual no-code and low-code studio, plus agentic computer vision |
| Model training | Not required | Optional, with tools to collect data and train custom models |
| Video inputs | IP and CCTV cameras (RTSP, ONVIF), NVRs, MP4 files, webcams, and cloud folders | Existing cameras, sensors, video systems, edge devices, and operational systems |
| Outputs | Slack, Microsoft Teams, email, direct API, and dashboards | Alerts, dashboards, automation, and ERP, BI, or third-party systems |
| Deployment | Hosted, self-serve web platform, with edge deployments on the Enterprise plan | Cloud, on premises, or edge, on a wide range of hardware |
| Getting started | Free plan and self-serve sign-up | Discovery and pilot with the viso.ai team |
| Pricing model | Usage credits shared across unlimited users | Custom, through the viso.ai sales team |
| Governance | SSO, SAML, and role-based access on Business, with audit logs on Enterprise | SOC 2 Type II, ISO 27001, SSO, RBAC, isolated tenants, full audit trail, and customer-controlled data |
| Best fit | Delivering use cases quickly with minimal IT involvement | Mission-critical vision at multi-site scale |
Viso Now and Viso Suite build agentic computer vision applications on the same VGI-1 engine. The choice comes down to who builds and owns the application, where it has to run, and how much integration and governance must surround it.
How Agentic Computer Vision Works in Both Products
Agentic computer vision describes systems that interpret a goal, decide which perception steps they need, reason over what the camera shows, and take an action such as sending an alert or writing a record. viso.ai’s overview of agentic computer vision covers the concept in depth, so this section focuses on how each product exposes it.
In Viso Now, the agent is the application builder. A safety manager can describe the outcome they want, for example flagging any person who enters a marked zone behind a reversing vehicle, and the agent assembles the input feed, detectors, logic, and alerts. The manager stays in control of every rule and threshold, which keeps a human in the loop for anything consequential.
In Viso Suite, agentic computer vision sits inside a wider platform that also supports custom models, edge runtimes, and enterprise workflows. That combination matters when an agent’s decision has to open a work order, feed a compliance report, or run on a device with no dependable cloud connection, and it gives the agent the deployment, monitoring, and governance needed to trust its output across hundreds of cameras and sites.
When To Choose Viso Now
The simplest rule for when to use Viso Now is speed: it is the stronger fit when learning quickly matters more than infrastructure, whether the buyer is a single site or a department inside a large enterprise. Typical signals include:
- You want to test a use case this week. A plain-language prompt replaces the data collection, annotation, and training cycle that usually delays a pilot.
- The people closest to the problem will build the app. Operations, EHS, quality, and facilities managers can create and refine applications without a machine learning team.
- Your cameras are already installed. Standard IP cameras, existing video recorders, and uploaded footage all work as inputs.
- Results need to reach existing tools. Alerts to Slack, Teams, or email fit how most frontline teams already communicate.
- Budget approval is still ahead of you. A free plan and shared credits make it possible to show evidence before asking for investment.
- Your company has governance rules. Business and Enterprise plans add SSO, SAML, role-based access, audit logs, and edge deployments, so enterprise teams can run Viso Now inside existing IT policies.
Innovation managers often use Viso Now to answer the question that stalls most computer vision programs, which is whether a use case works in their own environment. A short trial on real footage produces a sharper business case than a projection, and the Viso Now prompts guide offers ten operational starting points. For a worked example, viso.ai’s account of an equipment maker’s agentic vision pilot shows how one team scoped its trial and chose a no-code route.

When To Choose Viso Suite
The rule for when to use Viso Suite is just as clear: it is the stronger fit when a vision application becomes part of how the business runs. Typical signals include:
- The use case is mission-critical. A missed event carries safety, regulatory, or financial consequences, so reliability targets and support commitments matter.
- You operate many sites or many use cases. A single platform for building, deploying, and monitoring avoids a patchwork of point solutions.
- Data must stay in your infrastructure. Viso Suite keeps customer data within the customer’s environment and does not use visual customer data to train third-party models, as outlined in its privacy approach.
- Processing has to happen at the edge. Low latency, limited bandwidth, or privacy rules call for inference on the device.
- You need custom models or deep integrations. Specialized inspection tasks, ERP and BI connections, and custom workflows sit naturally in an end-to-end platform.
- Procurement asks for certifications. SOC 2 Type II, ISO/IEC 27001, and support for GDPR and the EU AI Act are baseline requirements for many enterprise buyers.
These are the conditions that cause promising pilots to stall in production, a pattern examined in why computer vision projects fail, and Viso Suite absorbs that complexity with a viso.ai team embedded from site audit to rollout.
Where Edge Computing Changes the Answer
Edge deployment, meaning video is processed on hardware at the site, is often the clearest dividing line between the two products. Processing video close to the cameras shortens response times, avoids sending large volumes of footage over the network, and keeps sensitive video on the premises, and each of those benefits matters more as an operation grows.
For a single site with good connectivity and non-urgent alerts, a hosted platform such as Viso Now is usually sufficient. When an application must react within a fraction of a second, run on a remote asset with intermittent connectivity, or keep footage on premises for regulatory reasons, edge computing for computer vision becomes a requirement, and Viso Suite’s hardware-agnostic deployment is built for that case. The line is not absolute, because the Viso Now Enterprise plan also lists edge deployments, so organizations with moderate edge needs should raise them early in either conversation.

Starting With Viso Now and Scaling With Viso Suite
Many organizations will not face an either-or decision. A common path uses each product for the stage it suits best:
- Explore. Teams test ideas in Viso Now against real footage and learn which questions their cameras can answer.
- Validate. The strongest use cases run live for several weeks, producing evidence of value and a clearer view of false positives and edge cases.
- Standardize. Proven applications are documented as requirements, including sites, cameras, integrations, and governance needs.
- Scale. Where those requirements call for enterprise deployment, edge processing, or custom models, the program moves into a Viso Suite discovery and pilot with the viso.ai team, while new ideas keep starting in Viso Now.

The Standardize step is the easiest one to skip. Writing down what a validated application needs keeps a successful trial from quietly becoming an unsupported production system, and it gives the scaling of AI vision a business case, success metrics, and internal champions from day one. Leadership teams that want a shared way to review AI risk before rollout can use the NIST AI Risk Management Framework, a voluntary guide for organizations that deploy or use AI. The guide to the cost of computer vision covers the budgeting side of that transition.

Questions To Ask Before You Decide
Five questions usually settle the choice, and the decision guide maps where each answer points:
- Who will build and maintain the application day to day?
- How many cameras, sites, and use cases will run within the next year?
- Must video or results stay inside your own infrastructure?
- Does any application need edge processing or a custom-trained model?
- Which certifications and governance controls will procurement require?
If most answers point toward a small team, a handful of cameras, and a need for quick evidence, start with Viso Now. If most point toward scale, data sovereignty, and integration, talk to the viso.ai team about Viso Suite.
