Block launches Berd desktop app for staff to work with AI agents across models under Apache 2.0 (Carl Franzen/VentureBeat)
agents
| Source: Techmeme | Original article
Block has launched Berd, an open‑source desktop app that lets its staff interact with AI agents from multiple models in a unified environment.
Block has open‑sourced a new desktop client called **Berd**, designed to give the company’s staff a unified workspace for interacting with AI agents built on a variety of underlying models. The tool, released under the permissive Apache 2.0 licence, bundles the interfaces for multiple model providers into a single application that runs on Windows, macOS and Linux. Block, the technology firm founded by former Twitter chief Jack Dorsey that now owns Square and Cash App, says the app was created to streamline internal experimentation and reduce the friction of switching between disparate AI platforms.
The move matters because it signals a shift from ad‑hoc, web‑based prompts toward more integrated development environments for AI‑driven workflows. By packaging cross‑model support into a lightweight desktop app, Block is lowering the barrier for engineers, product teams and analysts to prototype, test and iterate on agent behaviours without juggling separate APIs or browser tabs. The Apache 2.0 licence also invites external developers to fork, extend or embed Berd in their own stacks, potentially accelerating community‑driven innovation around multi‑model orchestration.
What to watch next is how quickly the broader developer community adopts Berd and whether other firms follow suit with similar open‑source toolkits. Block’s internal rollout may surface use‑case patterns that inform future updates, such as tighter security sandboxes, plug‑in ecosystems or integration with remote‑desktop solutions like AnyDesk or Chrome Remote Desktop for distributed teams. Observers will also be keen to see if the app’s architecture, which likely leans on lightweight frameworks such as Neutralinojs for cross‑platform delivery, proves scalable for larger enterprise AI deployments. The reception of Berd could shape the next wave of productivity tools that bridge the gap between disparate large‑language‑model services and everyday workplace tasks.
Sources
Back to AIPULSEN