audio-separator before 0.47.2 contains an insecure deserialization vulnerability that allows attackers to execute code by supplying malicious Demucs .th model files loaded via torch.load with weights_only=False. Local users writing to the shared /tmp/audio-separator-models/ default directory, or attackers controlling unverified model download sources, can execute code when victims run Demucs separation.
{
"cna_assigner": "VulnCheck",
"cwe_ids": [
"CWE-502"
],
"osv_generated_from": "https://github.com/CVEProject/cvelistV5/tree/main/cves/2026/134xxx/CVE-2026-134172.json"
}{
"extracted_events": [
{
"range": [
{
"introduced": "0"
},
{
"fixed": "0.47.2"
}
],
"source": "AFFECTED_FIELD",
"strategy": "StandardRange"
},
{
"range": [
{
"fixed": "6d7d3fd60accda60c7ad8807ce5731022475dcf3"
}
],
"source": "REFERENCES_COMMIT"
},
{
"original_tag": "v0.47.2",
"range": [
{
"fixed": "6d7d3fd60accda60c7ad8807ce5731022475dcf3"
}
],
"source": "REFERENCES_TAG"
}
]
}