CVE-2026-35030 — LiteLLM has an authentication bypass via OIDC userinfo cache key collision

Description (CNA)

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.

Affected (CNA)

Vendor / product	Versions
BerriAI / litellm	affected < 1.83.0

CWE

CWE-287

References

https://github.com/BerriAI/litellm/security/advisories/GHSA-jjhc-v7c2-5hh6

ADP enrichment

CISA-ADP

No English description supplied.

No references supplied.

redhat-SADP

A flaw was found in LiteLLM, a proxy server for Large Language Model (LLM) APIs. When JSON Web Token (JWT) authentication is enabled, the OIDC user information cache uses a truncated portion of the token as a cache key. An unauthenticated attacker can exploit this by crafting a JWT with the same initial characters as a legitimate user's cached token. This allows the attacker to bypass authentication and inherit the legitimate user's identity and permissions, potentially leading to unauthorized access and privilege escalation.