CVE-2026-100652
vLLM 0.22.0 through 0.23.0 Denial of Service via stop_token_ids
vLLM versions 0.22.0 through 0.23.0 fail to validate stop_token_ids against vocabulary bounds in Rust HTTP and gRPC frontends, allowing out-of-vocabulary token IDs to reach MinTokensLogitsProcessor. Attackers can submit requests with min_tokens greater than zero and out-of-vocabulary stop_token_ids to trigger CUDA tensor indexing failures that leave EngineCore in a fatal state requiring service restart.
Attack profile
How easy it is to exploit
- Attack vector
- Network
- Complexity
- High
- Privileges needed
- None
- User interaction
- None
What an attacker can affect
- Confidentiality
- None
- Integrity
- None
- Availability
- High
Exploit likelihood
What each source reports
CVE List Score used
vLLM 0.22.0 through 0.23.0 Denial of Service via stop_token_ids
vLLM versions 0.22.0 through 0.23.0 fail to validate stop_token_ids against vocabulary bounds in Rust HTTP and gRPC frontends, allowing out-of-vocabulary token IDs to reach MinTokensLogitsProcessor. Attackers can submit requests with min_tokens greater than zero and out-of-vocabulary stop_token_ids to trigger CUDA tensor indexing failures that leave EngineCore in a fatal state requiring service restart.
- Record
CVE-2026-100652- CVSS 4.0
- 8.2 High
CVSS:4.0/AV:N/AC:H/AT:P/PR:N/UI:N/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N - CVSS 3.1
- 5.9 Medium
CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:N/I:N/A:H - Weakness
- CWE-20
- Updated
- 2026-11-02
EUVDEUVD-2026-87746
vLLM versions 0.22.0 through 0.23.0 fail to validate stop_token_ids against vocabulary bounds in Rust HTTP and gRPC frontends, allowing out-of-vocabulary token IDs to reach MinTokensLogitsProcessor. Attackers can submit requests with min_tokens greater than zero and out-of-vocabulary stop_token_ids to trigger CUDA tensor indexing failures that leave EngineCore in a fatal state requiring service restart.
- Record
EUVD-2026-87746- CVSS 4.0
- 8.2 High
CVSS:4.0/AV:N/AC:H/AT:P/PR:N/UI:N/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N - Updated
- 2026-11-02
GitHub AdvisoriesGHSA-gxcm-7c4h-f56c
vLLM versions 0.22.0 through 0.23.0 fail to validate stop_token_ids against vocabulary bounds in...
vLLM versions 0.22.0 through 0.23.0 fail to validate stop_token_ids against vocabulary bounds in Rust HTTP and gRPC frontends, allowing out-of-vocabulary token IDs to reach MinTokensLogitsProcessor. Attackers can submit requests with min_tokens greater than zero and out-of-vocabulary stop_token_ids to trigger CUDA tensor indexing failures that leave EngineCore in a fatal state requiring service restart.
- Record
GHSA-gxcm-7c4h-f56c- CVSS 4.0
- 8.2 High
CVSS:4.0/AV:N/AC:H/AT:P/PR:N/UI:N/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N/E:X/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:X - CVSS 3.1
- 5.9 Medium
CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:N/I:N/A:H - Weakness
- CWE-20
- Updated
- 2026-11-02
NVD
vLLM versions 0.22.0 through 0.23.0 fail to validate stop_token_ids against vocabulary bounds in Rust HTTP and gRPC frontends, allowing out-of-vocabulary token IDs to reach MinTokensLogitsProcessor. Attackers can submit requests with min_tokens greater than zero and out-of-vocabulary stop_token_ids to trigger CUDA tensor indexing failures that leave EngineCore in a fatal state requiring service restart.
- Record
CVE-2026-100652- CVSS 4.0
- 8.2 High
CVSS:4.0/AV:N/AC:H/AT:P/PR:N/UI:N/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N/E:X/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:X - CVSS 3.1
- 5.9 Medium
CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:N/I:N/A:H - CVSS 3.1
- 7.5 High
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H - Weakness
- CWE-20
- Updated
- 2026-11-02
OSV
vLLM 0.22.0 through 0.23.0 Denial of Service via stop_token_ids
vLLM versions 0.22.0 through 0.23.0 fail to validate stop_token_ids against vocabulary bounds in Rust HTTP and gRPC frontends, allowing out-of-vocabulary token IDs to reach MinTokensLogitsProcessor. Attackers can submit requests with min_tokens greater than zero and out-of-vocabulary stop_token_ids to trigger CUDA tensor indexing failures that leave EngineCore in a fatal state requiring service restart.
- Record
CVE-2026-100652- CVSS 4.0
- 8.2 High
CVSS:4.0/AV:N/AC:H/AT:P/PR:N/UI:N/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N - Weakness
- CWE-20
- Updated
- 2026-11-02
OSVPYSEC-2026-4185
vLLM versions 0.22.0 through 0.23.0 fail to validate stop_token_ids against vocabulary bounds in Rust HTTP and gRPC frontends, allowing out-of-vocabulary token IDs to reach MinTokensLogitsProcessor. Attackers can submit requests with min_tokens greater than zero and out-of-vocabulary stop_token_ids to trigger CUDA tensor indexing failures that leave EngineCore in a fatal state requiring service restart.
- Record
PYSEC-2026-4185- CVSS 3.1
- 7.5 High
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H - Updated
- 2026-11-02
Affected products
Affected packages
| Package | Affected | Fixed |
|---|---|---|
vllmPyPI |
0.22.0 up to 0.24.0, plus listed versions 0.22.10.23.0 |
0.24.0 |
History
| Recorded | Changed | Sources |
|---|---|---|
| first indexed | CVE List, EUVD, GitHub Advisories, NVD, OSV |
Dates are when Every CVE recorded each change. Upstream dates may differ.
In short
CVE-2026-100652 is a high vulnerability (CVSS 8.2) in Vllm, published September 26, 2026. No source lists public exploit code. EPSS puts the chance of exploitation in the next 30 days at 0.42%, the 36th percentile. Fixed in vllm 0.24.0.
References
github.com/vllm-project/vllm/security/advisories/GHSA-qff2-492f-9fm4
vulncheck.com/advisories/vllm-0.22.0-through-0.23.0-denial-of-service-via-stop-token-ids
nvd.nist.gov/vuln/detail/CVE-2026-100652