CVE-2025-32375
published 2025-04-09CVE-2025-32375: BentoML is a Python library for building online serving systems optimized for AI apps and model inference. Prior to 1.4.8, there was an insecure…
PriorityP179critical9.8CVSS 3.1
AVNACLPRNUINSUCHIHAH
EXPLOIT
EPSS
45.77%
98.6th percentile
BentoML is a Python library for building online serving systems optimized for AI apps and model inference. Prior to 1.4.8, there was an insecure deserialization in BentoML's runner server. By setting specific headers and parameters in the POST request, it is possible to execute any unauthorized arbitrary code on the server, which will grant the attackers to have the initial access and information disclosure on the server. This vulnerability is fixed in 1.4.8.
Affected
4 ranges
| Vendor | Product | Version range | Fixed in |
|---|---|---|---|
| bentoml | bentoml | — | — |
| bentoml | bentoml | >= 1.0.0 < 1.4.8 | 1.4.8 |
| bentoml | bentoml | >= 1.0.0 < 1.4.8 | 1.4.8 |
| bentoml | bentoml | >= 1.0.0a1 < 1.4.8 | 1.4.8 |
Detection & IOCsextracted from sources · hover to see the quote
urlhttps://github.com/rapid7/metasploit-framework/blob/master/modules/exploits/linux/http/bentoml_runner_server_rce_cve_2025_32375.rb↗
- →Look for POST requests to BentoML runner server endpoints with suspicious or crafted headers that trigger deserialization — the exploit requires setting specific headers and parameters in the POST request to achieve RCE. ↗
- →A public Metasploit module exists for this CVE targeting Linux HTTP services running BentoML runner server — monitor for exploitation attempts matching the module's request patterns. ↗
- ·The vulnerability affects BentoML versions prior to 1.4.8 only; upgrading to 1.4.8 or later remediates the insecure deserialization in the runner server. ↗
CVEs like this are exactly what “Exploited This Week” covers.
Every Monday: what got weaponized or added to CISA KEV in the last seven days — each CVE cross-linked to its PoC, Nuclei template, and detection rule. Free, one email a week, unsubscribe in one click.
GHSA
BentoML's runner server Vulnerable to Remote Code Execution (RCE) via Insecure Deserialization
ghsa·2025-04-09
CVE-2025-32375 [CRITICAL] CWE-502 BentoML's runner server Vulnerable to Remote Code Execution (RCE) via Insecure Deserialization
BentoML's runner server Vulnerable to Remote Code Execution (RCE) via Insecure Deserialization
### Summary
There was an insecure deserialization in BentoML's runner server. By setting specific headers and parameters in the POST request, it is possible to execute any unauthorized arbitrary code on the server, which will grant the attackers to have the initial access and information disclosure on the server.
### PoC
- First, create a file named **model.py** to create a simple model and save it
```
import bentoml
import numpy as np
class mymodel:
def predict(self, info):
return np.abs(info)
def __call__(self, info):
return self.predict(info)
model = mymodel()
bentoml.picklable_model.save_model("mymodel", model)
```
- Then run the following command to save this model
```
python3 model.py
`
OSV
CVE-2025-32375: BentoML is a Python library for building online serving systems optimized for AI apps and model inference
osv·2025-04-09
CVE-2025-32375 CVE-2025-32375: BentoML is a Python library for building online serving systems optimized for AI apps and model inference
BentoML is a Python library for building online serving systems optimized for AI apps and model inference. Prior to 1.4.8, there was an insecure deserialization in BentoML's runner server. By setting specific headers and parameters in the POST request, it is possible to execute any unauthorized arbitrary code on the server, which will grant the attackers to have the initial access and information disclosure on the server. This vulnerability is fixed in 1.4.8.
OSV
BentoML's runner server Vulnerable to Remote Code Execution (RCE) via Insecure Deserialization
osv·2025-04-09
CVE-2025-32375 [CRITICAL] BentoML's runner server Vulnerable to Remote Code Execution (RCE) via Insecure Deserialization
BentoML's runner server Vulnerable to Remote Code Execution (RCE) via Insecure Deserialization
### Summary
There was an insecure deserialization in BentoML's runner server. By setting specific headers and parameters in the POST request, it is possible to execute any unauthorized arbitrary code on the server, which will grant the attackers to have the initial access and information disclosure on the server.
### PoC
- First, create a file named **model.py** to create a simple model and save it
```
import bentoml
import numpy as np
class mymodel:
def predict(self, info):
return np.abs(info)
def __call__(self, info):
return self.predict(info)
model = mymodel()
bentoml.picklable_model.save_model("mymodel", model)
```
- Then run the following command to save this model
```
python3 model.py
`
No detection rules found.
2025-04-09
Published