H Company Releases Runner H Public Beta Alongside Holo-1 and Tester H for Developers

H Company Releases Runner H Public Beta Alongside Holo-1 and Tester H for Developers

The concept behind Agentic AI is that multiple small, task-focused agents can collaborate to accomplish real work. H Company, based in Paris, aims to transform this concept into a practical reality. They have announced three significant advancements, starting with the public beta release of their flagship product, Runner H.

What is Runner H?

Runner H is an advanced AI agent designed to automate complex, multi-step tasks without the need for repetitive manual input. Essentially, it acts as a coordinator for your digital workload. Users provide a high-level goal, and Runner H’s internal orchestration system breaks it down into manageable tasks, intelligently assigning them to specialized sub-agents.

Key capabilities of Runner H include:

• Agent Orchestration: Users can initiate complete workflows using a single prompt, allowing Runner H to coordinate various AI agents to execute necessary steps in a synchronized manner.
• App Integration: It connects with popular tools such as Slack, Notion, Google Sheets, and various APIs, enabling immediate automation within existing software setups.
• Knowledge Uploads: Users can upload PDFs, documents, and data files, which Runner H processes to generate more accurate results.
• Autonomous Payments (Coming Soon): Future plans include enabling Runner H agents to make purchases from platforms like Amazon and Shopify without direct human intervention.

Runner H addresses common frustrations developers face with large language models (LLMs), such as fragmented context and weak execution. Instead of providing a single paragraph of advice, Runner H allocates tasks to mini-agents capable of planning, calling APIs, and executing tasks while maintaining context.

How to Use Runner H

To get started with Runner H:

1. Visit the Runner H web page on H Company’s website and click on Try Now.
2. Sign up to begin. Enter your prompt and submit.
3. Runner H will process the task, and you can track its progress from start to finish.

For example, you might prompt: “Visit [URL] and summarize the 3 latest AI research news articles.”

Holo-1: Open Weights for Visual Navigation

To enhance browser automation, H Company has released Holo-1, a family of action vision-language models with 3 billion and 7 billion parameters. When paired with Surfer H, Holo-1 achieves a score of 92.2% on the WebVoyager benchmark for UI localization while remaining efficient enough to run on a single GPU. Both the model weights and a 1,639-scenario WebClick dataset are available on Hugging Face, providing valuable resources for training agents capable of reading and interacting with web interfaces.

Tester H: Closing the QA Gap

As AI accelerates code generation, quality assurance processes must evolve. Tester H, currently in private beta for enterprises, converts plain-English user stories into executable end-to-end tests. This allows agents to navigate interfaces like real customers, clicking buttons, opening modals, and checking text and layout for any discrepancies.

Early tests indicate that Tester H has reduced regression cycles while increasing confidence in each code push. Requests to join the beta are open on H Company’s website.

Conclusion

H Company’s recent announcements, particularly the public beta launch of Runner H, signal a significant step toward integrating AI agents into everyday productivity. The introduction of Holo-1 and Tester H further enriches the AI landscape, addressing specific challenges in DevOps.

H Company anticipates that by 2025, AI agents will transition from being mere demonstrations to essential components of the toolchain. The ongoing development of sophisticated agentic AI capabilities is a trend worth monitoring in the evolving tech narrative.