CVE-2026-31251
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 gRPC server component. When the server starts, it loads the speech synthesis model from a user-specified directory using torch.load() without enabling the weights_only=True security parameter. This allows the deserialization of arbitrary Python objects via the pickle module. An attacker can exploit this by providing malicious model files within a directory. When a victim starts the gRPC server pointing to this directory, arbitrary code is executed on the victim's system during server initialization.
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's gRPC server loads speech synthesis models via torch.load() without the weights_only=True parameter, allowing deserialization of arbitrary Python objects through the pickle module. An attacker can place malicious model files in a directory; when a victim starts the gRPC server pointing to this directory, arbitrary code executes during model loading.
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
- Network
- Complexity
- Low
- Privileges
- None
- User interaction
- None
- Scope
- Unchanged
- Confidentiality
- Low
- Integrity
- Low
- Availability
- Low
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/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 via pip list. Also check if the cosyvoice package is present in your Python environment.Affected if CosyVoice is installed and the version matches the affected range (check against released versions containing the gRPC server functionality)
-
Locate gRPC server startup configurationSearch for gRPC server configuration files or scripts that start the CosyVoice gRPC service. Look for files like 'grpc_server.py', 'server.py', or configuration pointing to a gRPC port (typically 50051 or similar).Affected if gRPC server is enabled and configured to load models from a directory
-
Inspect torch.load() calls in model loading codeSearch the CosyVoice source code for 'torch.load(' calls. Inspect each call to verify if weights_only=True is set. Search using: grep -r 'torch.load' /path/to/cosyvoice/Affected if Any torch.load() call is found without weights_only=True parameter when loading models from user-specified directories
-
Verify model directory configurationCheck the gRPC server configuration or startup scripts to determine which directory is used for loading models. Look for arguments like '--model_dir', '--model_path', or environment variables pointing to model directories.Affected if The model directory is writable by other users or can be influenced by external parties
-
Check file permissions on model directoriesRun 'ls -la' on the configured model directory to verify if other users have write permissions, or check if models are loaded from shared/network locations.Affected if Model directory has weak permissions allowing another user to place malicious model files
You are affected if the CosyVoice gRPC server is running with models loaded via torch.load() without weights_only=True from a directory that could be compromised by an attacker.
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 when loading models from user-specified directories, and implement model file validation/checksumming to ensure only trusted models are loaded.
- Consultation3.0 h
- Implementation2.0 h
- Testing4.0 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 $3,072.
Scan for this in your stack
Free · runs locallyCheck whether your project pulls in CVE-2026-31251 — 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-2026-31251 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