CVE-2026-31252
Official description Straight from the sourceThe vendor's or NVD's own wording, published unedited. Authoritative, but often terse — it says what broke, rarely what to do.
NVD · uneditedCosyVoice thru commit 6e01309e01bc93bbeb83bdd996b1182a81aaf11e (2025-30-21) contains an insecure deserialization vulnerability (CWE-502) in its model loading component. The framework uses torch.load() to load model weight files (e.g., llm.pt, flow.pt, hift.pt) without enabling the security-restrictive weights_only=True parameter. This allows the deserialization of arbitrary Python objects via the pickle module. An attacker can exploit this by providing a malicious model directory containing specially crafted model files. When a victim starts the CosyVoice Web UI pointing to this directory, arbitrary code is executed on the victim's system during the model loading process.
Technical summary Written by usOur analysis, written from the advisory, the CVSS vector and the affected-version data. It adds context the advisory leaves out, and never invents facts that are not in the source.
dbcve analysis · high confidenceCosyVoice uses torch.load() to load model weight files (llm.pt, flow.pt, hift.pt) without the security-restrictive weights_only=True parameter, enabling arbitrary code execution via malicious pickle-serialized objects in model files when a victim loads them through the Web UI.
Verify against the referenced sources before acting — the references below are authoritative for this CVE, this summary is not.
CVSS breakdown How the score is builtThe industry scoring standard. It rates how the flaw is reached, what it takes to exploit, and what an attacker gains — the score is derived from those, not the other way round.
From the vector- Attack vector
- Local
- Complexity
- Low
- Privileges
- Low
- User interaction
- Required
- Scope
- Changed
- Confidentiality
- Low
- Integrity
- Low
- Availability
- Low
CVSS:3.1/AV:L/AC:L/PR:L/UI:R/S:C/C:L/I:L/A:L
Am I affected? How to checkSteps we derive from the advisory and the affected-version data, so you can decide whether this CVE reaches your setup. They are a guide, not a scan — your own configuration is the authority.
dbcve checksWork through these to decide whether this CVE applies to you.
-
Identify CosyVoice installation and versionRun 'pip show cosyvoice' or check the installed package version against the affected version range in the CVE documentationAffected if The installed version falls within the vulnerable version range and no patches have been applied
-
Locate torch.load() calls in CosyVoice source codeSearch the CosyVoice codebase for 'torch.load(' and examine each call for the presence of 'weights_only=True' parameterAffected if Any torch.load() call loads model files (llm.pt, flow.pt, hift.pt) without weights_only=True parameter
-
Verify Web UI component is enabledCheck if the CosyVoice Web UI service is running or configured to start, as this is the reported attack vectorAffected if The Web UI is active and can be used to trigger model file loading
-
Identify model file loading mechanismInspect how the Web UI loads model weight files - trace the code path from UI input to torch.load() callAffected if The Web UI directly passes user-supplied or fetched model files to torch.load() without weights_only=True
-
Check for model file integrity controlsSearch the codebase for any checksum verification, signature validation, or security checks performed on model files before torch.load() is calledAffected if No integrity verification exists before loading model files via torch.load()
You are affected if CosyVoice is installed with a vulnerable version, the codebase contains torch.load() calls for model files (llm.pt, flow.pt, hift.pt) without weights_only=True, and the Web UI can trigger loading of these files without integrity checks.
Generated from the published advisory. Verify against your own configuration.
Remediation Closing itWhat it takes to close this. Where a vendor fix exists we point at it; where none exists we say so plainly, and can build one. Effort estimates are scoped from the advisory, not from your codebase.
From vendor dataAdd weights_only=True to all torch.load() calls or migrate to a safe serialization format; implement model file integrity verification (e.g., checksums/signatures) before loading.
- Consultation2.0 h
- Implementation4.0 h
- Testing3.0 h
- Review / QA2.0 h
An estimate, not a bill — we confirm scope with you before any work starts. Need it this week? Rush from $3,088.
Scan for this in your stack
Free · runs locallyCheck whether your project pulls in CVE-2026-31252 — or any other known-vulnerable package — straight from your lock files. Free and open source; it runs locally and uploads nothing.
References Go to the primary sourcePrimary sources — vendor advisories, patches and trackers. Where our summary and a reference disagree, the reference wins.
Primary sourcesPractitioner notes
ContributedPeer-ranked notes from engineers who’ve handled CVE-2026-31252 in production — separate from our analysis above.
The advisory tells you what broke. It rarely tells you what actually worked. If you’ve dealt with this one, that detail is what the next engineer is searching for.
- The version that genuinely resolved it — not the one the vendor claimed
- A config change or rule that shut the vector down
- A gotcha in the upgrade path that cost you an afternoon
No notes yet
Be the first to add a field note for this CVE — a mitigation you’ve verified, a version caveat, or a link to a working fix. Sign in above to contribute.
A place for practitioners to share what actually worked: a mitigation you’ve tested, a configuration change, a version- or environment-specific caveat, or a link to a verified patch. The most useful notes rise to the top as peers upvote them, so the signal stays high.
- Verified mitigations, workarounds, and config changes
- Version or environment caveats, and links to real fixes
- No weaponised exploit code, or anything meant to cause harm
- No spam, self-promotion, credentials, or personal data