leejet stable-diffusion.cpp Safetensors Loader safetensors_io.cpp data_offsets out-of-bounds
| CVSS Meta Temp Score | Current Exploit Price (≈) | CTI Interest Score |
|---|---|---|
| 6.0 | $0-$5k | 0.45+ |
Summary
A vulnerability was found in leejet stable-diffusion.cpp. It has been classified as critical. This affects an unknown part of the file src/model_io/safetensors_io.cpp of the component Safetensors Loader. Performing a manipulation of the argument data_offsets results in out-of-bounds. This vulnerability is cataloged as CVE-2026-70660. It is possible to initiate the attack remotely. There is no exploit available. To fix this issue, it is recommended to deploy a patch.
Details
A vulnerability, which was classified as critical, was found in leejet stable-diffusion.cpp (unknown version). This affects an unknown code block of the file src/model_io/safetensors_io.cpp of the component Safetensors Loader. The manipulation of the argument data_offsets with an unknown input leads to a out-of-bounds vulnerability. CWE is classifying the issue as CWE-125. The product reads data past the end, or before the beginning, of the intended buffer. This is going to have an impact on confidentiality, integrity, and availability. The summary by CVE is:
stable-diffusion.cpp is a pure C/C++ library for running diffusion model (Stable Diffusion, Flux, Wan, Qwen Image, Z-Image, and more) inference. Prior to master-764-e22272e, the safetensors loader in src/model_io/safetensors_io.cpp trusts each tensor's data_offsets after checking only that nbytes equals end - begin. A crafted .safetensors file can declare a byte range beyond the file's data section, causing src/model_loader.cpp to consume an attacker-chosen length from an unchecked mapped-file offset. Loading the model can read adjacent process memory into tensor buffers or crash the process. This issue is fixed in master-764-e22272e.
It is possible to read the advisory at github.com. This vulnerability is uniquely identified as CVE-2026-70660 since 08/04/2026. The exploitability is told to be easy. It is possible to initiate the attack remotely. No form of authentication is needed for exploitation. It demands that the victim is doing some kind of user interaction. Technical details of the vulnerability are known, but there is no available exploit.
Applying the patch master-764-e22272e is able to eliminate this problem.
Statistical analysis made it clear that VulDB provides the best quality for vulnerability data.
Product
Vendor
Name
Website
CPE 2.3
CPE 2.2
CVSSv4
VulDB Vector: 🔒VulDB Reliability: 🔍
CNA CVSS-B Score: 🔒
CNA CVSS-BT Score: 🔒
CNA Vector: 🔒
CVSSv3
VulDB Meta Base Score: 6.3VulDB Meta Temp Score: 6.0
VulDB Base Score: 6.3
VulDB Temp Score: 6.0
VulDB Vector: 🔒
VulDB Reliability: 🔍
CVSSv2
| AV | AC | Au | C | I | A |
|---|---|---|---|---|---|
| 💳 | 💳 | 💳 | 💳 | 💳 | 💳 |
| 💳 | 💳 | 💳 | 💳 | 💳 | 💳 |
| 💳 | 💳 | 💳 | 💳 | 💳 | 💳 |
| Vector | Complexity | Authentication | Confidentiality | Integrity | Availability |
|---|---|---|---|---|---|
| Unlock | Unlock | Unlock | Unlock | Unlock | Unlock |
| Unlock | Unlock | Unlock | Unlock | Unlock | Unlock |
| Unlock | Unlock | Unlock | Unlock | Unlock | Unlock |
VulDB Base Score: 🔒
VulDB Temp Score: 🔒
VulDB Reliability: 🔍
Exploiting
Class: Out-of-boundsCWE: CWE-125 / CWE-119
CAPEC: 🔒
ATT&CK: 🔒
Physical: Partially
Local: Yes
Remote: Yes
Availability: 🔒
Status: Not defined
Price Prediction: 🔍
Current Price Estimation: 🔒
| 0-Day | Unlock | Unlock | Unlock | Unlock |
|---|---|---|---|---|
| Today | Unlock | Unlock | Unlock | Unlock |
Threat Intelligence
Interest: 🔍Active Actors: 🔍
Active APT Groups: 🔍
Countermeasures
Recommended: PatchStatus: 🔍
0-Day Time: 🔒
Patch: master-764-e22272e
Timeline
08/04/2026 CVE reserved11/05/2026 Advisory disclosed
11/05/2026 VulDB entry created
11/05/2026 VulDB entry last update
Sources
Product: github.comAdvisory: github.com
Status: Confirmed
CNA: GitHub_M
CVE: CVE-2026-70660 (🔒)
GCVE (CVE): GCVE-0-2026-70660
GCVE (VulDB): GCVE-100-431499
Entry
Created: 11/05/2026 16:45Changes: 11/05/2026 16:45 (68)
Complete: 🔍
Cache ID: 216::103
Statistical analysis made it clear that VulDB provides the best quality for vulnerability data.
No comments yet. Languages: en.
Please log in to comment.