What it is
CVE-2023-43654 is an unauthenticated TorchServe Server-Side Request Forgery affecting pytorch serve. TorchServe is a tool for serving and scaling PyTorch models in production. TorchServe default configuration lacks...
Vulnerability report
TorchServe Server-Side Request Forgery
pytorch / serve · affected before 0.8.2
Decision summary
Direct answers before the deeper technical record.
What it is
CVE-2023-43654 is an unauthenticated TorchServe Server-Side Request Forgery affecting pytorch serve. TorchServe is a tool for serving and scaling PyTorch models in production. TorchServe default configuration lacks...
Is it exploited?
Yes. KEV Intelligence tracks this CVE as a known exploited vulnerability. Confidence is high.
Who is affected?
pytorch / serve affected before 0.8.2.
What should we do?
Prioritize remediation. Validate affected assets and apply vendor fixes on an accelerated timeline.
Overview
TorchServe is a tool for serving and scaling PyTorch models in production. TorchServe default configuration lacks proper input validation, enabling third parties to invoke remote HTTP download requests and write files to the disk. This issue could be taken advantage of to compromise the integrity of the system and sensitive data. This issue is present in versions 0.1.0 to 0.8.1. A user is able to load the model of their choice from any URL that they would like to use.
The user of TorchServe is responsible for configuring both the allowed_urls and specifying the model URL to be used. A pull request to warn the user when the default value for allowed_urls is used has been merged in PR #2534. TorchServe release 0.8.2 includes this change. Users are advised to upgrade. There are no known workarounds for this issue.
Exploitation evidence
Third-party attestation and first-party sensor observation are shown separately so teams can judge the evidence chain.
The Shadowserver
Independent exploitation attestation added to the KEV Intelligence record.
GitHub
Public scanner or PoC coverage increases practical exploitability.
Per-source evidence links for KEV attestations are available through the KEV Intelligence Pro API.
Learn about Pro API access| Source | Added |
|---|---|
| The Shadowserver First | 2023-09-28 22:10 UTC |
Detection
Make the evidence actionable in scanner, SOC, and edge-control workflows.
Raw values available in Pro and Enterprise.
Metasploit template detected 28 Apr 2025.
View Metasploit template (opens in new tab)No KEV Intelligence virtual patch is currently available. Future rules ship for ModSecurity, Cloudflare, and AWS WAF.
Learn about virtual patches →Scanner and exploit-framework references linked to this CVE.
| Scanner | Reference | Detected |
|---|---|---|
| Metasploit | https://github.com/rapid7/metasploit-framework/blob/master/modules/exploits/multi/http/torchserver_cve_2023_43654.rb | 28 Apr 2025 |
| Nuclei | https://github.com/projectdiscovery/nuclei-templates/blob/main/http/cves/2023/CVE-2023-43654.yaml | 25 Apr 2025 |
Risk and context
CVSS v3.1
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:H/A:H
EPSS
35.3%
These PoCs are unverified and could contain malware. Use at your own risk.
nuclei · Created Unknown
Timeline
Public proof-of-concept code published
Exploit module available
Scanner coverage available
Vulnerability disclosed publicly
High-confidence, third-party attested exploitation
Identifier reserved by the CNA
Pro API
Confidence, exploit status, sensor telemetry, PoCs, scanner integrations, mentions, and tags are available programmatically for VM, SOC, and CTI workflows.
GET /api/v2/pro/kevs/CVE-2023-43654
Free JSON includes basic KEV fields{
"cve_id": "CVE-2023-43654",
"confidence": "High",
"cvss_score": 10.0,
"cvss_estimated": false,
"epss_score": 0.35256,
"exploit_status": {
"exploited_in_the_wild": true,
"active_exploitation_observed": false
},
"sensor_telemetry": { "attempts": 0, "sensors": 0 }
}
Early warning alerts
Receive curator-selected alerts when exploitation activity warrants attention. Each alert includes the evidence and context needed to decide what requires attention now.