Ray vulnerabilities
3 known CVE vulnerabilities in Ray, translated and rated.
- CVE-2025-62593HighActively exploited
Ray is an AI compute engine. Prior to version 2.52.0, developers using Ray as a development tool can be exploited via a critical RCE vulnerability exploitable via Firefox and Safari. This is due to insufficient guard against browser-based attacks, as the current defense uses the User-Agent header starting with 'Mozilla'. This defense is insufficient because the fetch specification allows modifying the User-Agent header. Combined with DNS rebinding, the vulnerability is exploitable against a developer who inadvertently visits a malicious website or sees a malicious ad. Patched in version 2.52.0.
- CVE-2026-57516High
Ray prior to 2.56.0 contains an unsafe deserialization vulnerability in the WebDataset reader. An attacker can achieve remote code execution by supplying a malicious tar archive to the read_webdataset() function, which unconditionally calls pickle.loads() on .pkl/.pickle entries and torch.load() with weights_only=False on .pt/.pth entries, executing arbitrary code on every remote worker processing the archive.
- CVE-2026-32981High
A path traversal vulnerability in Ray Dashboard (default port 8265) in versions prior to 2.8.1. Improper validation and sanitization of user-supplied paths in the static file handling mechanism allows an attacker to use traversal sequences (e.g., ../) to access files outside the intended static directory, resulting in local file disclosure.