CVE-2025-43849
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 · uneditedRetrieval-based-Voice-Conversion-WebUI is a voice changing framework based on VITS. Versions 2.2.231006 and prior are vulnerable to unsafe deserialization. The ckpt_a and cpkt_b variables take user input (e.g. a path to a model) and pass it to the merge function in process_ckpt.py, which uses them to load the models on those paths with torch.load, which can lead to unsafe deserialization and remote code execution. As of time of publication, no known patches exist.
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 confidenceRetrieval-based-Voice-Conversion-WebUI versions 2.2.231006 and prior contain an unsafe deserialization vulnerability in the checkpoint merging function. User-controlled input (ckpt_a and ckpt_b parameters specifying model file paths) is passed directly to torch.load() in process_ckpt.py without sanitization, allowing attackers to supply malicious pickle files that execute arbitrary code upon loading.
Verify against the referenced sources before acting — the references below are authoritative for this CVE, this summary is not.
Affected products & versions What the vendor confirmedThe version ranges the vendor confirmed as vulnerable. If your version sits inside a range here, treat yourself as exposed until you have upgraded.
NVD · CPE data<= 2.2.231006CVSS 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
- High
- Integrity
- High
- Availability
- High
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H
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.
-
Check RVC WebUI versionLocate the installed RVC WebUI directory and look for a version file, changelog, or check the git tag. Common locations include the root directory where the application is installed. Run: git describe --tags (if using git) or look for a version.py or similar file.Affected if The installed version is 2.2.231006 or any version prior to it.
-
Locate the vulnerable process_ckpt.py fileSearch for process_ckpt.py in the RVC WebUI installation directory. This file handles checkpoint merging functionality. Common path pattern: <install_dir>/infer/modules/voice_conversion/process_ckpt.py or similar location under the infer or modules directory.Affected if The file exists in the installation and is part of the deployed version.
-
Inspect torch.load() usage for unsafe deserializationOpen process_ckpt.py and search for torch.load() calls. Examine whether the ckpt_a and ckpt_b parameters (or similar model path parameters) are passed directly to torch.load() without sanitization. Look specifically for calls like torch.load(ckpt_a) or torch.load(request.form['ckpt_a']).Affected if torch.load() is called with user-supplied path input without weights_only=True or input validation.
-
Determine if checkpoint merging feature is exposedCheck if the web interface or API exposes a checkpoint merging endpoint. Look for routes in main.py, web.py, or similar files that handle /merge, /checkpoint/merge, or similar paths. Verify if this feature is enabled and accessible to users.Affected if The checkpoint merging feature is enabled and accessible without additional authentication or network restrictions.
-
Verify torch.load security flagIn process_ckpt.py, confirm that torch.load() does NOT use weights_only=True. If the call looks like torch.load(path) or torch.load(path, map_location=...), this is vulnerable. Only torch.load(path, weights_only=True) is secure.Affected if The torch.load() call lacks weights_only=True parameter, leaving deserialization unprotected.
You are affected if you run RVC WebUI version 2.2.231006 or earlier AND the checkpoint merging feature is accessible with process_ckpt.py containing unsafe torch.load() calls without weights_only=True.
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 dataUntil an official patch is released, restrict access to the checkpoint merging functionality to trusted users only. If the feature must remain exposed, add strict input validation and modify torch.load() calls to use weights_only=True or implement a safe loading wrapper.
- Consultation2.0 h
- Implementation4.0 h
- Testing3.0 h
- Review / QA1.0 h
An estimate, not a bill — we confirm scope with you before any work starts. Need it this week? Rush from $2,800.
Scan for this in your stack
Free · runs locallyCheck whether your project pulls in CVE-2025-43849 — 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-2025-43849 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