PatchSiren cyber security CVE debrief
CVE-2026-109225 sgl-project CVE debrief
CVE-2026-109225 is a high-severity vulnerability in SGLang versions up to 0.5.21, allowing unauthenticated remote attackers to cause a denial of service by exhausting memory through large file uploads to /v1/audio/transcriptions. Defenders should assess exposure, prioritize remediation, and verify system inventory. The vulnerability is caused by unbounded memory allocation, which can be triggered by concurrent large multipart uploads. This can lead to a denial of service, potentially causing system crashes or instability. Defenders should review system inventory, identify potentially affected versions, and update to a secure version if necessary. Additionally, defenders should be
- Vendor
- sgl-project
- Product
- sglang
- CVSS
- HIGH 8.7
- CISA KEV
- Not listed in stored evidence
- Original CVE published
- 2026-10-12
- Original CVE updated
- 2026-10-12
- Advisory published
- 2026-10-12
- Advisory updated
- 2026-10-12
Who should care
Defenders responsible for systems using SGLang should assess exposure and prioritize remediation. This includes reviewing system inventory, identifying potentially affected versions, and updating to a secure version if necessary.
Why it matters
CVE-2026-109225 is a high-severity vulnerability in SGLang that allows unauthenticated remote attackers to cause a denial of service. Defenders should assess exposure, prioritize remediation, and verify system inventory to mitigate potential impacts.
- Denial of service through memory exhaustion
- Potential system crashes or instability
- Need for verification of system inventory and exposure
- Remediation priority for affected systems
Technical summary
The vulnerability in SGLang through 0.5.21 allows unauthenticated remote attackers to cause a denial of service by uploading large files to /v1/audio/transcriptions, leading to memory exhaustion. The CVSS score for this vulnerability is 8.7, indicating a high severity. This is caused by unbounded memory allocation which can be triggered by concurrent large multipart uploads that are fully read into memory before validation to crash the HTTP/tokenizer worker and deny service. Defenders should assess exposure, prioritize remediation, and verify system inventory to mitigate potential impacts. The vulnerability affects SGLang versions up to 0.5.21 and can be mitigated by updating to a secure version.
Defensive priority
Remediation is recommended for systems using SGLang version 0.5.21 or earlier. Verify system inventory and assess exposure to CVE-2026-109225.
Recommended defensive actions
- Assess exposure by verifying system inventory for SGLang version 0.5.21 or earlier
- Prioritize remediation for affected systems
- Monitor system resources for signs of potential memory exhaustion attacks
- Review compensating controls for exposed systems while remediation is scheduled and verified
- Check relevant monitoring, detection, and logs for exposed assets that need extra review
- Track exceptions, retest remediated assets, and close the item only after evidence is documented
- Confirm whether affected product deployments exist in managed environments and assign an owner for follow-up
Evidence notes
The CVE record and source item provide details on the vulnerability, including its description, CVSS score, and affected versions. However, the corpus does not establish specific exploitation instances or remediation steps beyond version updates.
Sources and references
Verified primary and authoritative sources
-
CVE-2026-109225 CVE Program record
Publisher, destination, and source semantics verified
URL: https://www.cve.org/CVERecord?id=CVE-2026-109225
CVE Program - Official CVE Program record with source-provided CVE metadata.
-
CVE-2026-109225 NVD vulnerability detail
Publisher, destination, and source semantics verified
URL: https://nvd.nist.gov/vuln/detail/CVE-2026-109225
NIST National Vulnerability Database - Official NIST NVD detail page and source-specific vulnerability assessment.
Supplemental references
-
SGLang through 0.5.21 Memory Exhaustion DoS via /v1/audio/transcriptions Upload
Unverified legacy reference
URL: https://raw.githubusercontent.com/CVEProject/cvelistV5/main/cves/2026/109xxx/CVE-2026-109225.json
cve_program_cvelist_v5
-
Source reference
Unverified legacy reference
URL: https://github.com/sgl-project/sglang/issues/43709
Supplemental source - issue-tracking
-
Source reference
Unverified legacy reference
URL: https://github.com/sgl-project/sglang/blob/v0.5.21/python/sglang/srt/entrypoints/http_server.py
Supplemental source - technical-description
-
Source reference
Unverified legacy reference
URL: https://github.com/sgl-project/sglang
Supplemental source - product
-
Source reference
Unverified legacy reference
URL: https://www.vulncheck.com/advisories/sglang-through-0.5.21-memory-exhaustion-dos-via-v1-audio-transcriptions-upload
Supplemental source - third-party-advisory
Methodology and review provenance
AI-assisted synthesis based on stored public vulnerability evidence. System validation, approval state, and publication status do not by themselves establish human review of this revision. PatchSiren helps prioritize defensive review and does not prove exposure or remediation on any system.