# CVE-2025-70999: OneFlow denial of service in flow.cuda.get_device_capability

Source: https://junglewise.ai/threats/cve-2025-70999-oneflow-denial-of-service-in-flow-cuda-get-device-capability  
Publisher: Junglewise Threat Intelligence, https://junglewise.ai/threats  
Published: 2026-01-28. Updated: 2026-07-05  
License: CC BY 4.0 with a link back

CVE-2025-70999 · Severity: high · CVSS 7.5 · Published 2026-01-28

OneFlow is an open-source deep learning framework used for training and deploying artificial intelligence models. A flaw in how the software handles GPU device identifiers allows an attacker to crash the application by providing an invalid ID. This results in a denial-of-service, potentially disrupting AI training workloads or production inference services.

Technologies: [Oneflow](https://junglewise.ai/threats/technologies/oneflow). Vendors: [Oneflow](https://junglewise.ai/threats/vendors/oneflow).

## Quick answers

**Is CVE-2025-70999 exploited in the wild?** No. Junglewise Threat Intelligence has no report of CVE-2025-70999 being exploited in the wild.

**How severe is CVE-2025-70999?** High severity, with a CVSS score of 7.5, published 28 January 2026.

**Which products and versions are affected?** OneFlow-Inc OneFlow 0.9.0.

**Is there a fix for CVE-2025-70999?** No fixed version is recorded yet. Check the vendor's advisory in the references below. If you run it, Junglewise drafts a mitigation that fits your setup and applies it once you approve, with senior security engineers behind it.

**How is CVE-2025-70999 attacked?** It is attacked over the network, in a software library.

**Am I affected by CVE-2025-70999?** Check whether you run Oneflow. Junglewise does that check for you around the clock: connect your code, cloud and domains, and it matches new vulnerabilities like this one against what you actually run and tells you when one reaches you.

## Technical details

A vulnerability exists in OneFlow v0.9.0 within the `flow.cuda.get_device_capability()` function due to insufficient validation of the `device` parameter. When an invalid or out-of-bounds device ID is provided, the underlying `cudaSetDevice()` call fails, triggering a `Check failed` assertion in `cuda_util.cpp`. This leads to an immediate process crash (SIGABRT) and core dump. An attacker capable of influencing the device ID passed to this function can remotely or locally cause a denial-of-service (DoS) condition. The issue is documented in the project's GitHub issue tracker as a bug where the framework fails to gracefully handle invalid device ordinals.

## Affected products

- OneFlow-Inc OneFlow 0.9.0

## Timeline

- 2025-06-30: disclosed: Issue reported on GitHub repository
- 2026-01-28: advisory: CVE published to NVD

## References

- [https://github.com/Daisy2ang](https://github.com/Daisy2ang)
- [https://github.com/Oneflow-Inc/oneflow/issues/10660](https://github.com/Oneflow-Inc/oneflow/issues/10660)

## Related threats

- [Oneflow-Inc OneFlow segmentation fault in flow.column_stack](https://junglewise.ai/threats/cve-2025-71001-oneflow-inc-oneflow-segmentation-fault-in-flow-column-stack)
- [OneFlow flow.cuda.BoolTensor denial of service via crafted input](https://junglewise.ai/threats/cve-2025-71000-oneflow-flow-cuda-booltensor-denial-of-service-via-crafted-input)
- [OneFlow GPU device-ID validation denial of service in get_device_properties](https://junglewise.ai/threats/cve-2025-65891-oneflow-gpu-device-id-validation-denial-of-service-in-get-device)
- [OneFlow denial of service via invalid device ID in CUDA synchronization](https://junglewise.ai/threats/cve-2025-65890-oneflow-denial-of-service-via-invalid-device-id-in-cuda)
- [Oneflow-Inc OneFlow denial of service in flow.dstack](https://junglewise.ai/threats/cve-2025-65889-oneflow-inc-oneflow-denial-of-service-in-flow-dstack)