CVE-2026-10300: Reachable Assertion in SGLang
CVE-2026-10300 is a medium severity vulnerability in SGLang version 0. 5. 10. post1 involving a reachable assertion triggered by manipulation of the lora_path argument in the file python/sglang/srt/lora/lora_manager. py within the Inference HTTP Endpoint component. The vulnerability can be exploited remotely but requires high attack complexity and no user interaction. Although the exploit has been publicly disclosed, no known exploits are currently observed in the wild. A fix has been proposed via a pull request but has not yet been accepted or officially released.
AI Analysis
Technical Summary
This vulnerability in SGLang 0.5.10.post1 arises from a reachable assertion triggered by crafted input to the lora_path argument in the Inference HTTP Endpoint's lora_manager.py file. The vulnerability allows remote attackers to cause an assertion failure, potentially disrupting service or causing denial of service. The attack complexity is high, and no privileges or user interaction are required. The vulnerability has a CVSS 4.0 base score of 6.3, indicating medium severity. A patch is pending acceptance but not yet officially available.
Potential Impact
The vulnerability can lead to assertion failures remotely, which may cause denial of service or application instability. Due to the high complexity of exploitation and lack of known active exploits, the immediate risk is moderate. No privilege escalation, confidentiality, or integrity impacts are indicated.
Mitigation Recommendations
Currently, no official patch or fix is available as the pull request to address this issue awaits acceptance. Users should monitor the vendor or project repository for the official patch release. Until then, consider restricting access to the vulnerable endpoint if feasible to reduce exposure.
CVE-2026-10300: Reachable Assertion in SGLang
Description
CVE-2026-10300 is a medium severity vulnerability in SGLang version 0. 5. 10. post1 involving a reachable assertion triggered by manipulation of the lora_path argument in the file python/sglang/srt/lora/lora_manager. py within the Inference HTTP Endpoint component. The vulnerability can be exploited remotely but requires high attack complexity and no user interaction. Although the exploit has been publicly disclosed, no known exploits are currently observed in the wild. A fix has been proposed via a pull request but has not yet been accepted or officially released.
CVSS v4.0
Score 6.3medium
AI-Powered Analysis
Machine-generated threat intelligence
Technical Analysis
This vulnerability in SGLang 0.5.10.post1 arises from a reachable assertion triggered by crafted input to the lora_path argument in the Inference HTTP Endpoint's lora_manager.py file. The vulnerability allows remote attackers to cause an assertion failure, potentially disrupting service or causing denial of service. The attack complexity is high, and no privileges or user interaction are required. The vulnerability has a CVSS 4.0 base score of 6.3, indicating medium severity. A patch is pending acceptance but not yet officially available.
Potential Impact
The vulnerability can lead to assertion failures remotely, which may cause denial of service or application instability. Due to the high complexity of exploitation and lack of known active exploits, the immediate risk is moderate. No privilege escalation, confidentiality, or integrity impacts are indicated.
Mitigation Recommendations
Currently, no official patch or fix is available as the pull request to address this issue awaits acceptance. Users should monitor the vendor or project repository for the official patch release. Until then, consider restricting access to the vulnerable endpoint if feasible to reduce exposure.
Technical Details
- Data Version
- 5.2
- Assigner Short Name
- VulDB
- Date Reserved
- 2026-05-31T18:11:11.981Z
- Cvss Version
- 4.0
- State
- PUBLISHED
- Remediation Level
- null
Threat ID: 6a1e134de29bf47b5055d5c4
Added to database: 6/1/2026, 11:18:37 PM
Last enriched: 6/1/2026, 11:34:29 PM
Last updated: 6/2/2026, 6:42:52 AM
Views: 9
Community Reviews
0 reviewsCrowdsource mitigation strategies, share intel context, and vote on the most helpful responses. Sign in to add your voice and help keep defenders ahead.
Want to contribute mitigation steps or threat intel context? Sign in or create an account to join the community discussion.
Actions
Updates to AI analysis require Pro Console access. Upgrade inside Console → Billing.
Need more coverage?
Upgrade to Pro Console for AI refresh and higher limits.
For incident response and remediation, OffSeq services can help resolve threats faster.
Latest Threats
Check if your credentials are on the dark web
Instant breach scanning across billions of leaked records. Free tier available.