CVE-2026-35030: LiteLLM · Authentication bypass via OIDC userinfo cache key collision
Affects litellmSource GCVE Database
Not known to be exploited · CVSS 9.4 Critical : upgrade to 1.83.0 as a priority.
Data as of 16 Nov 2026, 15:20 UTC
- Patch
- Available – fixed in 1.83.0
CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:N/VC:H/VI:H/VA:N/SC:H/SI:H/SA:NExploitation evidence
Disclosed: evidence found.
CVE published 6 Apr 2026
Detectable: no evidence on record.
No public nuclei template indexed
Weaponized: no evidence on record.
No public exploit code indexed in Metasploit or Exploit-DB
Exploited in the wildExploited: no evidence on record.
Not in CISA KEV and no source reports exploitation
Observed by OffSeq MirageSeen by Mirage: no evidence on record.
No attack traffic recorded by OffSeq Mirage sensors
A missing step is not evidence of safety. Sources: cve.org (opens in a new tab), the Metasploit, Exploit-DB and nuclei-templates index as of 15 Nov 2026.
What to do
Upgrade litellm to 1.83.0.
Patches and advisories
Affected software
| Product | Affected | Fixed in |
|---|---|---|
| PyPIlitellm | Affected
| Fixed in
|
Check your own hosts for these packages with threat-finder (opens in a new tab), our free open-source scanner.
AI-suggested
Apply the official patch by upgrading LiteLLM to version 1.83.0 or later, which fixes the cache key collision issue by using the full hash of the JWT token. Alternatively, disable OIDC userinfo caching by setting the cache TTL to 0 or disable JWT authentication entirely. Since a patch is available, remediation should prioritize updating to the fixed version.
Description
LiteLLM versions prior to 1.83.0 with JWT authentication enabled are vulnerable to an authentication bypass due to OIDC userinfo cache key collisions. The cache key uses only the first 20 characters of the JWT token, which can be identical for different tokens signed with the same algorithm. An attacker can exploit this to impersonate legitimate users by crafting tokens with matching cache keys. This vulnerability is fixed in version 1.83.0 by using the full hash of the JWT token as the cache key. The vulnerability has a critical severity with a CVSS score of 9.4. Most instances are not affected as JWT authentication is not enabled by default.
References
- github.com/BerriAI/litellm/security/advisories/GHSA-jjhc-v7c2-5hh6 (opens in a new tab)
- nvd.nist.gov/vuln/detail/CVE-2026-35030 (opens in a new tab)
- access.redhat.com/errata/RHSA-2026:13545 (opens in a new tab)
- access.redhat.com/errata/RHSA-2026:28960 (opens in a new tab)
- access.redhat.com/errata/RHSA-2026:30056 (opens in a new tab)
- access.redhat.com/security/cve/CVE-2026-35030 (opens in a new tab)
- bugzilla.redhat.com/show_bug.cgi?id=2455509 (opens in a new tab)
- github.com/BerriAI/litellm (opens in a new tab)
- github.com/pypa/advisory-database/tree/main/vulns/litellm/PYSEC-2026-390.yaml (opens in a new tab)
- pypi.org/project/litellm (opens in a new tab)
- security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-35030.json (opens in a new tab)