CVE-2026-111160
7.5 HIGHpublic exploit availablePublished 2026-10-15 · Updated 2026-10-15
AI analysis for this CVE has not been generated yet. Raw NVD data is shown below.
NVD description
LightLLM through 1.2.0 contains a denial of service vulnerability that allows unauthenticated attackers to crash the inference worker by supplying the internal pd_kv_trans_params field to the /generate endpoint. Attackers can send a single request whose parameters cause pickle.loads to fail in InferSamplingParams, killing the inference thread while /health stays up and all later requests hang until restart.
CVSS vector
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H
Weaknesses
CWE-502
Public exploit & PoC references
- https://github.com/ModelTC/LightLLM
- https://github.com/ModelTC/LightLLM/issues/1629
- https://github.com/ModelTC/lightllm/blob/v1.2.0/lightllm/server/api_lightllm.py#L37-L43
- https://github.com/ModelTC/lightllm/blob/v1.2.0/lightllm/server/core/objs/sampling_params.py#L356-L357
- https://github.com/ModelTC/lightllm/blob/v1.2.0/lightllm/server/router/model_infer/infer_batch.py#L484-L485
All references
- https://github.com/ModelTC/LightLLM
- https://github.com/ModelTC/LightLLM/issues/1629
- https://github.com/ModelTC/lightllm/blob/v1.2.0/lightllm/server/api_lightllm.py#L37-L43
- https://github.com/ModelTC/lightllm/blob/v1.2.0/lightllm/server/core/objs/sampling_params.py#L356-L357
- https://github.com/ModelTC/lightllm/blob/v1.2.0/lightllm/server/router/model_infer/infer_batch.py#L484-L485
- https://www.vulncheck.com/advisories/lightllm-through-1.2.0-unauthenticated-dos-via-pd-kv-trans-params-deserialization
Source data: NVD (nvd.nist.gov), public domain. Exploit-DB.ai adds local AI analysis for defensive use only.
Related CVEs
- CRITICALCVE-2017-20285PoC
- HIGHCVE-2019-25777PoC
- CRITICALCVE-2025-59953PoC
- UNSCOREDCVE-2025-64393
- CRITICALCVE-2025-66455PoC
- HIGHCVE-2026-100214
- HIGHCVE-2026-100308PoC
- HIGHCVE-2026-100506
Related by shared AI tags and CWE weakness class. Browse the full archive.