Unauthenticated code execution via scheduler pickle deserialization
Published May 18, 2026 · Updated May 18, 2026
Unsafe deserialization in SGLang 0.5.5 through 0.5.12 allows remote attackers to execute code through the multimodal scheduler socket. The scheduler reads the final frame of each multipart ZeroMQ message and passes it directly to pickle.loads(), allowing attacker-selected reducers to run. Exploitation requires the multimodal runtime and network access to the ROUTER socket when --host 0.0.0.0 exposes it.
Summary
What happened
Unsafe deserialization in SGLang 0.5.5 through 0.5.12 allows remote attackers to execute code through the multimodal scheduler socket. The scheduler reads the final frame of each multipart ZeroMQ message and passes it directly to pickle.loads(), allowing attacker-selected reducers to run. Exploitation requires the multimodal runtime and network access to the ROUTER socket when --host 0.0.0.0 exposes it.
The record
- CVE
- CVE-2026-7301
- Published
- May 18, 2026
- Updated
- May 18, 2026
- Vendor
- SGLang
- Product
- SGLang
- Classifications
- CWE-502, T1190
- Attack vector
- network
- Privileges
- unauthenticated
Timeline
How it unfolded
- May 18, 2026CVE publishedPublication date reported by the CVE source.
- May 18, 2026Record updatedLatest update available in the CVE record.
Exploitability
Present is not the same as exploitable
Compare your product and version with the public record. A matching version still requires validation against your environment.
Is a vulnerable build present?
Compare these published version ranges with your installed build and any vendor patches.
- Affected versionversion=5.10
What conditions does exploitation require?
What is affected?
Attacks
What attackers are doing with it
Daily unique IPs observed by Shadowserver honeypots for known exploited vulnerabilities (KEVs). Missing observations do not establish an absence of attacks.
Weakness, pattern, technique
Public exploit references
- Antiproof scheduler pickle payload demonstrationproof of concept · demonstrated
Labels summarize the accepted research assessment. They do not indicate a test against your environment.
Technologies
Your stack
See the directory against your own environment.
Your stack
Check the software in your environment
Book a demo to see how Hinoki identifies affected software and validates exploitability in your environment.
Book a demo