CVE-2026-76850 (GCVE-0-2026-76850)
Vulnerability from cvelistv5 – Published: 2026-08-19 21:41 – Updated: 2026-08-19 21:41
VLAI
EPSS
VEX
Title
LMDeploy Remote Code Execution via Unsafe Pickle Deserialization in the Disaggregated Serving Peer Connector
Summary
LMDeploy deserializes disaggregated-serving peer messages with pickle. The handle_zmq_recv coroutine in lmdeploy/pytorch/disagg/conn/engine_conn.py reads peer-to-peer cache-free requests with recv_pyobj(), which deserializes the received bytes with pickle.loads(), and the isinstance check against DistServeCacheFreeRequest runs only after deserialization has already completed. The peer that supplies those bytes is caller-controlled: p2p_connect passes remote_engine_endpoint_info.zmq_address from the request body to connect() on the ZMQ PULL socket, and the POST /distserve/p2p_initialize and /distserve/p2p_connect endpoints in lmdeploy/serve/openai/api_server.py apply no authentication unless the server is started with api_keys, which defaults to None. A remote attacker can direct an engine to pull from a ZMQ endpoint under their control and execute arbitrary code in the engine process. Deployments that do not enable disaggregated serving are not affected, because the receive loop is only started once the migration backend accepts the connection.
Severity
9.8 (Critical)
CWE
- CWE-502 - Deserialization of Untrusted Data
Assigner
References
7 references
| URL | Tags |
|---|---|
| https://github.com/InternLM/lmdeploy/issues/4804 | issue-tracking |
| https://github.com/InternLM/lmdeploy/commit/f05b4… | patch |
| https://github.com/InternLM/lmdeploy/blob/v0.15.0… | technical-description |
| https://github.com/InternLM/lmdeploy/blob/v0.15.0… | technical-description |
| https://github.com/InternLM/lmdeploy/releases/tag… | release-notes |
| https://github.com/InternLM/lmdeploy | product |
| https://www.vulncheck.com/advisories/lmdeploy-rem… | third-party-advisory |
Impacted products
Date Public
2026-07-29 00:00
{
"containers": {
"cna": {
"affected": [
{
"collectionURL": "https://pypi.org",
"defaultStatus": "unaffected",
"packageName": "lmdeploy",
"packageURL": "pkg:pypi/lmdeploy",
"product": "lmdeploy",
"repo": "https://github.com/InternLM/lmdeploy",
"vendor": "InternLM",
"versions": [
{
"lessThan": "0.16.0",
"status": "affected",
"version": "0.9.2",
"versionType": "semver"
}
]
}
],
"cpeApplicability": [
{
"nodes": [
{
"cpeMatch": [
{
"criteria": "cpe:2.3:a:internlm:lmdeploy:*:*:*:*:*:*:*:*",
"versionEndExcluding": "0.16.0",
"versionStartIncluding": "0.9.2",
"vulnerable": true
}
],
"negate": false,
"operator": "OR"
}
]
}
],
"credits": [
{
"lang": "en",
"type": "reporter",
"value": "AAtomical"
}
],
"datePublic": "2026-07-29T00:00:00.000Z",
"descriptions": [
{
"lang": "en",
"value": "LMDeploy deserializes disaggregated-serving peer messages with pickle. The handle_zmq_recv coroutine in lmdeploy/pytorch/disagg/conn/engine_conn.py reads peer-to-peer cache-free requests with recv_pyobj(), which deserializes the received bytes with pickle.loads(), and the isinstance check against DistServeCacheFreeRequest runs only after deserialization has already completed. The peer that supplies those bytes is caller-controlled: p2p_connect passes remote_engine_endpoint_info.zmq_address from the request body to connect() on the ZMQ PULL socket, and the POST /distserve/p2p_initialize and /distserve/p2p_connect endpoints in lmdeploy/serve/openai/api_server.py apply no authentication unless the server is started with api_keys, which defaults to None. A remote attacker can direct an engine to pull from a ZMQ endpoint under their control and execute arbitrary code in the engine process. Deployments that do not enable disaggregated serving are not affected, because the receive loop is only started once the migration backend accepts the connection."
}
],
"metrics": [
{
"cvssV4_0": {
"attackComplexity": "LOW",
"attackRequirements": "NONE",
"attackVector": "NETWORK",
"baseScore": 9.3,
"baseSeverity": "CRITICAL",
"privilegesRequired": "NONE",
"subAvailabilityImpact": "NONE",
"subConfidentialityImpact": "NONE",
"subIntegrityImpact": "NONE",
"userInteraction": "NONE",
"vectorString": "CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N",
"version": "4.0",
"vulnAvailabilityImpact": "HIGH",
"vulnConfidentialityImpact": "HIGH",
"vulnIntegrityImpact": "HIGH"
},
"format": "CVSS"
},
{
"cvssV3_1": {
"attackComplexity": "LOW",
"attackVector": "NETWORK",
"availabilityImpact": "HIGH",
"baseScore": 9.8,
"baseSeverity": "CRITICAL",
"confidentialityImpact": "HIGH",
"integrityImpact": "HIGH",
"privilegesRequired": "NONE",
"scope": "UNCHANGED",
"userInteraction": "NONE",
"vectorString": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H",
"version": "3.1"
},
"format": "CVSS"
}
],
"problemTypes": [
{
"descriptions": [
{
"cweId": "CWE-502",
"description": "Deserialization of Untrusted Data",
"lang": "en",
"type": "CWE"
}
]
}
],
"providerMetadata": {
"dateUpdated": "2026-08-19T21:41:28.068Z",
"orgId": "83251b91-4cc7-4094-a5c7-464a1b83ea10",
"shortName": "VulnCheck"
},
"references": [
{
"name": "GitHub Issue #4804",
"tags": [
"issue-tracking"
],
"url": "https://github.com/InternLM/lmdeploy/issues/4804"
},
{
"name": "fix(disagg): use JSON instead of pickle for P2P ZMQ requests",
"tags": [
"patch"
],
"url": "https://github.com/InternLM/lmdeploy/commit/f05b4ad8bf2e2d84101a1d63b3c44fadd99223b2"
},
{
"name": "Vulnerable recv_pyobj() call in v0.15.0",
"tags": [
"technical-description"
],
"url": "https://github.com/InternLM/lmdeploy/blob/v0.15.0/lmdeploy/pytorch/disagg/conn/engine_conn.py#L79"
},
{
"name": "Caller-controlled zmq_address passed to connect() in v0.15.0",
"tags": [
"technical-description"
],
"url": "https://github.com/InternLM/lmdeploy/blob/v0.15.0/lmdeploy/pytorch/disagg/conn/engine_conn.py#L61"
},
{
"name": "lmdeploy v0.16.0",
"tags": [
"release-notes"
],
"url": "https://github.com/InternLM/lmdeploy/releases/tag/v0.16.0"
},
{
"tags": [
"product"
],
"url": "https://github.com/InternLM/lmdeploy"
},
{
"name": "VulnCheck Advisory: LMDeploy Remote Code Execution via Unsafe Pickle Deserialization in the Disaggregated Serving Peer Connector",
"tags": [
"third-party-advisory"
],
"url": "https://www.vulncheck.com/advisories/lmdeploy-remote-code-execution-via-unsafe-pickle-deserialization-in-the-disaggregated-serving-peer-connector"
}
],
"title": "LMDeploy Remote Code Execution via Unsafe Pickle Deserialization in the Disaggregated Serving Peer Connector",
"x_generator": {
"engine": "vulncheck-endgame"
}
}
},
"cveMetadata": {
"assignerOrgId": "83251b91-4cc7-4094-a5c7-464a1b83ea10",
"assignerShortName": "VulnCheck",
"cveId": "CVE-2026-76850",
"datePublished": "2026-08-19T21:41:28.068Z",
"dateReserved": "2026-08-19T21:24:37.594Z",
"dateUpdated": "2026-08-19T21:41:28.068Z",
"state": "PUBLISHED"
},
"dataType": "CVE_RECORD",
"dataVersion": "5.2",
"vulnerability-lookup:meta": {
"nvd": "{\"cve\":{\"id\":\"CVE-2026-76850\",\"sourceIdentifier\":\"disclosure@vulncheck.com\",\"published\":\"2026-08-19T22:17:28.123\",\"lastModified\":\"2026-08-19T22:17:28.123\",\"vulnStatus\":\"Received\",\"cveTags\":[],\"descriptions\":[{\"lang\":\"en\",\"value\":\"LMDeploy deserializes disaggregated-serving peer messages with pickle. The handle_zmq_recv coroutine in lmdeploy/pytorch/disagg/conn/engine_conn.py reads peer-to-peer cache-free requests with recv_pyobj(), which deserializes the received bytes with pickle.loads(), and the isinstance check against DistServeCacheFreeRequest runs only after deserialization has already completed. The peer that supplies those bytes is caller-controlled: p2p_connect passes remote_engine_endpoint_info.zmq_address from the request body to connect() on the ZMQ PULL socket, and the POST /distserve/p2p_initialize and /distserve/p2p_connect endpoints in lmdeploy/serve/openai/api_server.py apply no authentication unless the server is started with api_keys, which defaults to None. A remote attacker can direct an engine to pull from a ZMQ endpoint under their control and execute arbitrary code in the engine process. Deployments that do not enable disaggregated serving are not affected, because the receive loop is only started once the migration backend accepts the connection.\"}],\"affected\":[{\"source\":\"disclosure@vulncheck.com\",\"affectedData\":[{\"vendor\":\"InternLM\",\"product\":\"lmdeploy\",\"defaultStatus\":\"unaffected\",\"collectionURL\":\"https://pypi.org\",\"packageName\":\"lmdeploy\",\"repo\":\"https://github.com/InternLM/lmdeploy\",\"packageURL\":\"pkg:pypi/lmdeploy\",\"versions\":[{\"version\":\"0.9.2\",\"lessThan\":\"0.16.0\",\"versionType\":\"semver\",\"status\":\"affected\"}]}]}],\"metrics\":{\"cvssMetricV40\":[{\"source\":\"disclosure@vulncheck.com\",\"type\":\"Secondary\",\"cvssData\":{\"version\":\"4.0\",\"vectorString\":\"CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:H/VI:H/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\",\"baseScore\":9.3,\"baseSeverity\":\"CRITICAL\",\"attackVector\":\"NETWORK\",\"attackComplexity\":\"LOW\",\"attackRequirements\":\"NONE\",\"privilegesRequired\":\"NONE\",\"userInteraction\":\"NONE\",\"vulnConfidentialityImpact\":\"HIGH\",\"vulnIntegrityImpact\":\"HIGH\",\"vulnAvailabilityImpact\":\"HIGH\",\"subConfidentialityImpact\":\"NONE\",\"subIntegrityImpact\":\"NONE\",\"subAvailabilityImpact\":\"NONE\",\"exploitMaturity\":\"NOT_DEFINED\",\"confidentialityRequirement\":\"NOT_DEFINED\",\"integrityRequirement\":\"NOT_DEFINED\",\"availabilityRequirement\":\"NOT_DEFINED\",\"modifiedAttackVector\":\"NOT_DEFINED\",\"modifiedAttackComplexity\":\"NOT_DEFINED\",\"modifiedAttackRequirements\":\"NOT_DEFINED\",\"modifiedPrivilegesRequired\":\"NOT_DEFINED\",\"modifiedUserInteraction\":\"NOT_DEFINED\",\"modifiedVulnConfidentialityImpact\":\"NOT_DEFINED\",\"modifiedVulnIntegrityImpact\":\"NOT_DEFINED\",\"modifiedVulnAvailabilityImpact\":\"NOT_DEFINED\",\"modifiedSubConfidentialityImpact\":\"NOT_DEFINED\",\"modifiedSubIntegrityImpact\":\"NOT_DEFINED\",\"modifiedSubAvailabilityImpact\":\"NOT_DEFINED\",\"Safety\":\"NOT_DEFINED\",\"Automatable\":\"NOT_DEFINED\",\"Recovery\":\"NOT_DEFINED\",\"valueDensity\":\"NOT_DEFINED\",\"vulnerabilityResponseEffort\":\"NOT_DEFINED\",\"providerUrgency\":\"NOT_DEFINED\"}}],\"cvssMetricV31\":[{\"source\":\"disclosure@vulncheck.com\",\"type\":\"Primary\",\"cvssData\":{\"version\":\"3.1\",\"vectorString\":\"CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H\",\"baseScore\":9.8,\"baseSeverity\":\"CRITICAL\",\"attackVector\":\"NETWORK\",\"attackComplexity\":\"LOW\",\"privilegesRequired\":\"NONE\",\"userInteraction\":\"NONE\",\"scope\":\"UNCHANGED\",\"confidentialityImpact\":\"HIGH\",\"integrityImpact\":\"HIGH\",\"availabilityImpact\":\"HIGH\"},\"exploitabilityScore\":3.9,\"impactScore\":5.9}]},\"weaknesses\":[{\"source\":\"disclosure@vulncheck.com\",\"type\":\"Primary\",\"description\":[{\"lang\":\"en\",\"value\":\"CWE-502\"}]}],\"references\":[{\"url\":\"https://github.com/InternLM/lmdeploy\",\"source\":\"disclosure@vulncheck.com\"},{\"url\":\"https://github.com/InternLM/lmdeploy/blob/v0.15.0/lmdeploy/pytorch/disagg/conn/engine_conn.py#L61\",\"source\":\"disclosure@vulncheck.com\"},{\"url\":\"https://github.com/InternLM/lmdeploy/blob/v0.15.0/lmdeploy/pytorch/disagg/conn/engine_conn.py#L79\",\"source\":\"disclosure@vulncheck.com\"},{\"url\":\"https://github.com/InternLM/lmdeploy/commit/f05b4ad8bf2e2d84101a1d63b3c44fadd99223b2\",\"source\":\"disclosure@vulncheck.com\"},{\"url\":\"https://github.com/InternLM/lmdeploy/issues/4804\",\"source\":\"disclosure@vulncheck.com\"},{\"url\":\"https://github.com/InternLM/lmdeploy/releases/tag/v0.16.0\",\"source\":\"disclosure@vulncheck.com\"},{\"url\":\"https://www.vulncheck.com/advisories/lmdeploy-remote-code-execution-via-unsafe-pickle-deserialization-in-the-disaggregated-serving-peer-connector\",\"source\":\"disclosure@vulncheck.com\"}]}}"
}
}
Loading…
Loading…
Experimental. This forecast is provided for visualization only and may change without notice. Do not use it for operational decisions.
Forecast uses a logistic model when the trend is rising, or an exponential decay model when the trend is falling. Fitted via linearized least squares.
Sightings
| Author | Source | Type | Date | Other |
|---|
Nomenclature
- Seen: The vulnerability was mentioned, discussed, or observed by the user.
- Confirmed: The vulnerability has been validated from an analyst's perspective.
- Published Proof of Concept: A public proof of concept is available for this vulnerability.
- Exploited: The vulnerability was observed as exploited by the user who reported the sighting.
- Patched: The vulnerability was observed as successfully patched by the user who reported the sighting.
- Not exploited: The vulnerability was not observed as exploited by the user who reported the sighting.
- Not confirmed: The user expressed doubt about the validity of the vulnerability.
- Not patched: The vulnerability was not observed as successfully patched by the user who reported the sighting.
Loading…
The MITRE ATT&CK techniques below are AI-generated suggestions, inferred from the description of the
vulnerability by the CIRCL/vulnerability-attack-technique-classification-roberta-base
model, served locally by ML-Gateway.
They have not been verified by an analyst and are provided for guidance only.
The approach is described in our paper Mapping CVEs to MITRE ATT&CK Techniques: A Curated Gold-Set Classifier and the Limits of LLM-Assisted Label Expansion.
The approach is described in our paper Mapping CVEs to MITRE ATT&CK Techniques: A Curated Gold-Set Classifier and the Limits of LLM-Assisted Label Expansion.
Loading…
Loading…