LMDeploy has Remote Code Execution by Pickle Deserialization via handle_zmq_recv in lmdeploy/lmdeploy/pytorch/disagg/conn/engine_conn.py
LMDeploy is a toolkit for compressing, deploying, and serving large language models. Starting in version 0.9.2 and prior to version 0.16.0, LMDeploy's PyTorch DistServe/PD-disaggregation control plane used `recv_pyobj()` to deserialize messages received through a ZeroMQ PULL socket. PyZMQ implements `recv_pyobj()` using Python pickle deserialization, which can execute arbitrary code while reconstructing an object. The peer address used by the receiver was supplied through the `POST /distserve/p2p_connect` HTTP endpoint. An attacker who could reach an affected DistServe API server could cause the server to connect to an attacker-controlled ZeroMQ endpoint and deserialize a crafted pickle payload. API-key authentication is not enabled unless the operator explicitly configures it. As a result, affected DistServe deployments without API keys allowed unauthenticated remote code execution with the privileges of the LMDeploy serving process. This issue affects the PyTorch backend when PD-disaggregation/DistServe is enabled. Ordinary deployments that do not use the affected disaggregated-serving path do not expose this data flow. The fix was released in LMDeploy 0.16.0. Users who cannot upgrade immediately should prevent untrusted clients from reaching `/distserve/*` endpoints, restrict the DistServe HTTP and ZeroMQ control planes to trusted cluster networks, configure API-key authentication, and block arbitrary outbound ZeroMQ connections from serving nodes. These measures reduce ex
- Products
- Not specified
- Affected versions
- Not specified
- Fixed versions
- Not specified
- CVEs
- CVE-2025-66455
- Confirmed fixed CVEs
- Not specified
- Updated in this feed
- 2 Nov 2026, 23:50 UTC · Revision 1
Source evidence
LMDeploy has Remote Code Execution by Pickle Deserialization via handle_zmq_recv in lmdeploy/lmdeploy/pytorch/disagg/conn/engine_conn.py
core-osv · Original sourcePublished 18 Sept 2026, 17:03 UTC
LMDeploy is a toolkit for compressing, deploying, and serving large language models. Starting in version 0.9.2 and prior to version 0.16.0, LMDeploy's PyTorch DistServe/PD-disaggregation control plane used `recv_pyobj()` to deserialize messages received through a ZeroMQ PULL socket. PyZMQ implements `recv_pyobj()` using Python pickle deserialization, which can execute arbitrary code while reconstructing an object. The peer address used by the receiver was supplied through the `POST /distserve/p2p_connect` HTTP endpoint. An attacker who could reach an affected DistServe API server could cause the server to connect to an attacker-controlled ZeroMQ endpoint and deserialize a crafted pickle payload. API-key authentication is not enabled unless the operator explicitly configures it. As a result, affected DistServe deployments without API keys allowed unauthenticated remote code execution with the privileges of the LMDeploy serving process. This issue affects the PyTorch backend when PD-disaggregation/DistServe is enabled. Ordinary deployments that do not use the affected disaggregated-serving path do not expose this data flow. The fix was released in LMDeploy 0.16.0. Users who cannot upgrade immediately should prevent untrusted clients from reaching `/distserve/*` endpoints, restrict the DistServe HTTP and ZeroMQ control planes to trusted cluster networks, configure API-key authentication, and block arbitrary outbound ZeroMQ connections from serving nodes. These measures reduce ex
osv:CVE-2025-66455
Complete source record
{
"id": "6cfc8b865224dd9a1869664e92df4e52851d91ad9a57e7a84ae43284546e5f20",
"url": "https://osv.dev/vulnerability/CVE-2025-66455",
"method": "structured",
"payload": {
"kind": "advisory",
"title": "LMDeploy has Remote Code Execution by Pickle Deserialization via handle_zmq_recv in lmdeploy/lmdeploy/pytorch/disagg/conn/engine_conn.py",
"summary": "LMDeploy is a toolkit for compressing, deploying, and serving large language models. Starting in version 0.9.2 and prior to version 0.16.0, LMDeploy's PyTorch DistServe/PD-disaggregation control plane used `recv_pyobj()` to deserialize messages received through a ZeroMQ PULL socket. PyZMQ implements `recv_pyobj()` using Python pickle deserialization, which can execute arbitrary code while reconstructing an object. The peer address used by the receiver was supplied through the `POST /distserve/p2p_connect` HTTP endpoint. An attacker who could reach an affected DistServe API server could cause the server to connect to an attacker-controlled ZeroMQ endpoint and deserialize a crafted pickle payload. API-key authentication is not enabled unless the operator explicitly configures it. As a result, affected DistServe deployments without API keys allowed unauthenticated remote code execution with the privileges of the LMDeploy serving process. This issue affects the PyTorch backend when PD-disaggregation/DistServe is enabled. Ordinary deployments that do not use the affected disaggregated-serving path do not expose this data flow. The fix was released in LMDeploy 0.16.0. Users who cannot upgrade immediately should prevent untrusted clients from reaching `/distserve/*` endpoints, restrict the DistServe HTTP and ZeroMQ control planes to trusted cluster networks, configure API-key authentication, and block arbitrary outbound ZeroMQ connections from serving nodes. These measures reduce ex",
"evidence": "osv:CVE-2025-66455",
"products": [],
"severity": "critical",
"structured": {
"id": "CVE-2025-66455",
"aliases": [
"CVE-2025-66455",
"GHSA-2vh9-42vm-xmv2",
"PYSEC-2026-4077"
],
"metrics": [
{
"type": "CVSS_V3",
"score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H"
}
],
"affected": [
{
"ranges": [
{
"repo": "https://github.com/internlm/lmdeploy",
"type": "GIT",
"events": [
{
"introduced": "9098ae8c1abf35fdf81fb279d466c12766973aa7"
},
{
"fixed": "1208bf006bbac69f1f012ceafeeeb70f623b632c"
},
{
"fixed": "f05b4ad8bf2e2d84101a1d63b3c44fadd99223b2"
}
],
"database_specific": {
"extracted_events": [
{
"range": [
{
"introduced": "0.9.2"
},
{
"fixed": "0.16.0"
}
],
"source": "AFFECTED_FIELD",
"strategy": "StringRangeExpression"
},
{
"range": [
{
"fixed": "f05b4ad8bf2e2d84101a1d63b3c44fadd99223b2"
}
],
"source": "REFERENCES_COMMIT"
},
{
"range": [
{
"fixed": "1208bf006bbac69f1f012ceafeeeb70f623b632c"
}
],
"source": "REFERENCES_TAG",
"original_tag": "v0.16.0"
}
]
}
}
],
"versions": [
"v0.14.0",
"v0.15.0",
"0.14.0a2",
"v0.13.0",
"v0.12.3",
"v0.12.2",
"v0.12.1",
"v0.12.0",
"v0.11.1",
"v0.11.0",
"v0.10.2",
"v0.10.1",
"v0.10.0",
"v0.9.2"
],
"database_specific": {
"source": "https://storage.googleapis.com/cve-osv-conversion/osv-output/CVE-2025-66455.json"
}
}
],
"modified": "2026-10-28T02:30:24.544287888Z",
"provider": "osv",
"references": [
"https://github.com/InternLM/lmdeploy/releases/tag/v0.16.0",
"https://github.com/CVEProject/cvelistV5/tree/main/cves/2025/66xxx/CVE-2025-66455.json",
"https://github.com/InternLM/lmdeploy/security/advisories/GHSA-2vh9-42vm-xmv2",
"https://nvd.nist.gov/vuln/detail/CVE-2025-66455",
"https://github.com/InternLM/lmdeploy/commit/f05b4ad8bf2e2d84101a1d63b3c44fadd99223b2"
],
"provider_metadata": {
"cwe_ids": [
"CWE-502"
],
"cna_assigner": "GitHub_M",
"osv_generated_from": "https://github.com/CVEProject/cvelistV5/tree/main/cves/2025/66xxx/CVE-2025-66455.json"
}
},
"related_cves": [
{
"value": "CVE-2025-66455",
"evidence": "/id"
}
],
"security_relevant": true,
"severity_evidence": "/severity/0/score"
},
"source_id": "core-osv",
"document_id": 490992,
"published_at": "2026-09-18T17:03:37.084+00:00"
}