CVE-2026-10300 SGLang Inference HTTP Endpoint lora_manager.py assertion
Remote attackers can trigger a reachable assertion in the SGLang inference endpoint, causing a denial of service.
- Exploited in the wild
- Not known
- Login needed
- Not needed
- Attackable
- Over the internet
- Victim action needed
- Not needed
- Fix
- None listed
What is CVE-2026-10300?
CVE-2026-10300 is a medium-severity denial of service vulnerability in SGLang, published on 1 Jun 2026. An attacker can exploit it over the internet without logging in, with no action from the victim.
Official description
A security vulnerability has been detected in SGLang 0.5.10.post1. Impacted is an unknown function of the file python/sglang/srt/lora/lora_manager.py of the component Inference HTTP Endpoint. Such manipulation of the argument lora_path leads to reachable assertion. The attack can be launched remotely. A high complexity level is associated with this attack. The exploitability is considered difficult. The exploit has been disclosed publicly and may be used. The pull request to fix this issue awaits acceptance.
Weakness: CWE-617
Which versions are affected by CVE-2026-10300?
3 open source packages ship an affected version, including sglang.
| Package | Affected | Fixed in |
|---|---|---|
sglang PyPI | All versions to 0.5.10.post1 | None |
sglang PyPI | All versions to 0.5.10.post1 | None |
sglang PyPI | All versions to 0.5.10.post1 | None |
How do I fix CVE-2026-10300?
No fix has been published for CVE-2026-10300 yet. Limit exposure of the affected software until one is available.
Upgrade, then put limits on request size, rate and nesting depth so one client cannot exhaust the service. More about Denial of service.
Is CVE-2026-10300 being exploited?
It is not confirmed as exploited in the wild, but public exploit code for CVE-2026-10300 exists.
Sources
Also known as GHSA-m2jr-x4gq-5rmj, PYSEC-2026-3064.