Description
A vulnerability was detected in turboderp ExLlamaV2 up to 0.3.2. Affected is the function STFile.get_tensor of the file exllamav2/stloader.py of the component Safetensors Tensor Loader. Performing a manipulation of the argument data_offsets results in buffer overflow. The attack is possible to be carried out remotely. The exploit is now public and may be used. The project was informed of the problem early through an issue report but has not responded yet.
Affected products
| Vendor | Product | Affected | Fixed |
|---|---|---|---|
| turboderp | ExLlamaV2 | 0.3.0, 0.3.1, 0.3.2 | — |
References
- VDB-419848 | turboderp ExLlamaV2 Safetensors Tensor Loader stloader.py STFile.get_tensor buffer overflow vdb-entry technical-description
- VDB-419848 | CTI Indicators (IOB, IOC, IOA) signature permissions-required
- CVE-2026-110695 | CVE Analysis and Report third-party-advisory
- Submit #1003958 | turboderp ExLlamaV2 0.3.2 (commit 6a2d8311) and master commit 7dc12af Heap-based Buffer Overflow third-party-advisory
- https://github.com/turboderp-org/exllamav2/issues/820 issue-tracking
- https://github.com/3em0/cve_repo/blob/main/2026/ExLlamaV2-0.3.2-Heap-Buffer-Overflow-in-the-safetensors-tensor-loader(STFile.get_tensor-stloader_read).md exploit