{"id":94335,"date":"2025-05-07T04:02:14","date_gmt":"2025-05-07T04:02:14","guid":{"rendered":"https:\/\/neclink.com\/index.php\/2025\/05\/07\/hugging-face-releases-a-free-operator-like-agentic-ai-tool\/"},"modified":"2025-05-07T04:02:14","modified_gmt":"2025-05-07T04:02:14","slug":"hugging-face-releases-a-free-operator-like-agentic-ai-tool","status":"publish","type":"post","link":"https:\/\/neclink.com\/index.php\/2025\/05\/07\/hugging-face-releases-a-free-operator-like-agentic-ai-tool\/","title":{"rendered":"Hugging Face releases a free Operator-like agentic AI tool"},"content":{"rendered":"


\n<\/p>\n

\n

A team at Hugging Face has released a freely available, cloud-hosted computer-using AI \u201cagent.\u201d But be forewarned: It\u2019s quite sluggish and occasionally makes mistakes.<\/p>\n

Hugging Face\u2019s agent, called Open Computer Agent<\/a>, is accessible via the web and can use a Linux virtual machine preloaded with several applications, including Firefox. Similar to OpenAI\u2019s Operator<\/a>, you can prompt Open Computer Agent to complete a task \u2014 say, \u201cUse Google Maps to find the Hugging Face HQ in Paris\u201d \u2014 and sit back as the agent opens the necessary programs and figures out the required steps.<\/p>\n

Open Computer Agent can handle simple requests well enough. But more complicated ones, like searching for flights, tripped it up in TechCrunch\u2019s testing. Open Computer Agent also often runs into CAPTCHA tests that it\u2019s unable to solve. <\/p>\n

You\u2019ll also have to wait in a virtual queue to use Open Computer Agent \u2014 a queue seconds to minutes long, depending on demand.<\/p>\n

\n
\n
\n

We’re launching Computer Use in smolagents! \ud83e\udd73<\/p>\n

-> As vision models become more capable, they become able to power complex agentic workflows. Especially Qwen-VL models, that support built-in grounding, i.e. ability to locate any element in an image by its coordinates, thus to\u2026 pic.twitter.com\/mI8MuWZkIS<\/a><\/p>\n

\u2014 m_ric (@AymericRoucher) May 6, 2025<\/a><\/p><\/blockquote>\n<\/div>\n<\/figure>\n

Of course, the Hugging Face team\u2019s goal wasn\u2019t to build a state-of-the-art computer-using agent. Rather, they wanted to demonstrate that open AI models are becoming more capable \u2014 and cheaper to run on cloud infrastructure.<\/p>\n

\u201cAs vision models become more capable, they become able to power complex agentic workflows,\u201d Aymeric\u00a0Roucher, a member of the agents team at Hugging Face, wrote in a post on X<\/a>. \u201c[Some of these models] support built-in grounding, i.e. [the] ability to locate any element in an image by its coordinates, [and] thus [can] click any item [in a virtual machine].\u201d<\/p>\n

While it\u2019s far from perfect, agentic technology is attracting increasing investment as enterprises look to adopt it to boost productivity. According to a recent KPMG survey<\/a>, 65% of companies are experimenting with AI agents. Markets and Markets projects<\/a> that the AI agent segment will grow from $7.84 billion in 2025 to $52.62 billion by 2030.<\/p>\n

\n
\n

Techcrunch event<\/p>\n

\n

\n\t\t\t\t\t\t\t\t\tBerkeley, CA<\/span>
\n\t\t\t\t\t\t\t\t\t\t\t\t\t|<\/span>
\n\t\t\t\t\t\t\t\t\t\t\t\t\tJune 5<\/span>\n\t\t\t\t\t\t\t<\/p>\n

\t\t\t\t\t\t\t
\n\t\t\t\t\tBOOK NOW<\/span>
\n\t\t\t\t<\/a>\n\t\t\t\t\t<\/div>\n<\/p><\/div>\n<\/div>\n<\/div>\n