QX Labs builds AI agents, Grids and Flows for sales, research and ops teams. Free plan with 5,000 credits; Pro from $24/month billed annually.
Researched and fact-checked by Bartu Cavusoglu · Updated October 26, 2026
QX, made by QX Labs, is a platform for building AI agents and automations that operate inside the tools a team already uses. The company positions it as a way to delegate work rather than chat with an assistant: agents connect to a team's apps, carry out tasks and return finished output in Slack, Teams, WhatsApp or email, without requiring technical expertise.
The product is organized around three building blocks: Agents, Grids and Flows. A fourth element, Knowledge Vaults, indexes internal documents so answers stay grounded in company material. QX says it connects to more than 1,000 external apps. Its site names investors, consultants, sales, marketing and recruiting teams as typical users, particularly those who research and act on data daily.
Agents, Grids and Flows are all built through chat. A user describes the agent, spreadsheet-style run or recurring task, and QX sets it up. Agents are autonomous language models with custom prompts, skills, tools and data. According to the site, they can draft emails, update records, produce files and control their own computer. Teams can brief them like a colleague, for example a Research Analyst or Account Manager.
Grids run many agents in parallel across rows, so the same logic applies to thousands of records such as a lead list. Flows turn a described recurring task into a workflow that runs on a schedule, on a trigger or on request. Knowledge sources such as SharePoint, Google Drive, Notion and Granola sync into a shared base, with answers cited back to source.
Beyond the three core primitives, QX lets teams choose among models from OpenAI, Anthropic and Google Gemini, and bring their own API keys. Permission controls define what an agent may do in a connected tool such as Gmail, with some actions, like sending or deleting emails, blocked or requiring human approval. Access can be limited to one person, specific people or the whole organization.
Every run is traceable. Users can inspect inputs, outputs, credit usage and each step, and validate results on a sample before scaling up. On security, the site lists ISO 27001 certification, GDPR alignment and CASA Tier 2 certification, with SOC 2 Type II in progress. It states that customer data is not used to train models.
QX is a platform for building AI agents, Grids and Flows that connect to a team's tools and complete work such as research, enrichment and reporting. It works inside Slack, Teams, WhatsApp and email and connects to more than 1,000 apps.
The Free plan costs $0 and includes 5,000 credits per month and 3 seats. Pro is listed at $24 per month billed annually at $288, with 12k credits per month and up to 15 seats. Enterprise pricing is custom.
Yes. The Free plan requires no credit card and includes all features, scheduled and triggered workflows and 1,000+ integrations. It is limited to 5,000 credits per month and 3 seats, and top-ups are not available.
Existing grids, flows and agents stay accessible, but new executions cannot run until the next monthly reset, an upgrade to a larger package or a top-up. Top-ups cost $2.50 per 1,000 credits.
QX is built for teams that work with data daily, including investment firms, consultancies, sales, marketing and recruiting teams. It targets high-volume, repetitive tasks such as research, enrichment and process automation.
QX supports major providers including OpenAI, Anthropic and Google Gemini, and users can choose a model per task. Users can also bring their own API keys.
QX lists ISO 27001 certification, GDPR alignment and CASA Tier 2 certification, with SOC 2 Type II in progress. It states that data is encrypted in transit and at rest and that it does not train models on customer data.
No. The company says anyone comfortable with ChatGPT or Excel can go from idea to a working automation in minutes. Agents, grids and workflows are built through chat.