Vulnerability Description
LiteLLM is a proxy server (AI Gateway) to call LLM APIs in OpenAI (or native) format. Prior to 1.83.0, when JWT authentication is enabled (enable_jwt_auth: true), the OIDC userinfo cache uses token[:20] as the cache key. JWT headers produced by the same signing algorithm generate identical first 20 characters. This configuration option is not enabled by default. Most instances are not affected. An unauthenticated attacker can craft a token whose first 20 characters match a legitimate user's cached token. On cache hit, the attacker inherits the legitimate user's identity and permissions. This affects deployments with JWT/OIDC authentication enabled. Fixed in v1.83.0.
CVSS Score
CRITICAL
Affected Products
| Vendor | Product | Versions |
|---|---|---|
| Litellm | Litellm | < 1.83.0 |
Related Weaknesses (CWE)
References
- https://github.com/BerriAI/litellm/security/advisories/GHSA-jjhc-v7c2-5hh6Vendor AdvisoryMitigation
FAQ
What is CVE-2026-35030?
CVE-2026-35030 is a vulnerability with a CVSS score of 9.1 (CRITICAL). LiteLLM is a proxy server (AI Gateway) to call LLM APIs in OpenAI (or native) format. Prior to 1.83.0, when JWT authentication is enabled (enable_jwt_auth: true), the OIDC userinfo cache uses token[:2...
How severe is CVE-2026-35030?
CVE-2026-35030 has been rated CRITICAL with a CVSS base score of 9.1/10. This is considered a critical vulnerability requiring immediate attention.
Is there a patch for CVE-2026-35030?
Check the references section above for vendor advisories and patch information. Affected products include: Litellm Litellm.