CVE-2026-115919Medium· 6.2▾ SunlitONNX: Out-of-Bounds Read in RNN/GRU/LSTM Shape Inference (Old Opsets, Missing Rank Check)
▾ Sunlit zone — Low / medium · no exploitation signal
impact 34.1 · likelihood 0 · exploitation 0
Need a working PoC? Pro members can cast a request and our team develops one — it lands right here.
Stakeholder-Specific Vulnerability Categorization from CISA's ADP record at CVE.org: whether exploitation is observed, whether an attack can be automated, and how much of the system is at stake.
Exploit-prediction probability, daily snapshots since Oct 27.
Disclosure to exploitation, from the record and what we observed since indexing it.
0.2%
Disclosed via OSV
ONNX shape inference performed an out-of-bounds read when processing a malformed RNN, GRU, or LSTM node from opsets 1–6 whose first input had a rank other than three.
RNNShapeInference_opset1_to_6() checked whether the input shape was present but accessed dim(0) and dim(1) without first checking its rank. A crafted model could therefore terminate a process that called onnx.shape_inference.infer_shapes(). The model passed onnx.checker.check_model() before reaching the vulnerable path.
The vulnerable code read dimensions unconditionally:
if (hasInputShape(ctx, 0)) {
const auto& first_input_shape = getInputShape(ctx, 0);
seq_length = first_input_shape.dim(0);
batch_size = first_input_shape.dim(1);
}
For an undersized shape, protobuf's repeated-field access read outside the intended array in release builds. The observed result was SIGSEGV; sanitizer and debug builds terminated on a protobuf bounds check.
The newer opset 7–13 and opset 14+ inference paths already required the first input to have rank three and were not affected.
An application that performs shape inference on an untrusted model can be terminated, causing denial of service. Exploitation does not require privileges or human interaction once the model is processed. No confidentiality or integrity impact was demonstrated.
The vulnerable schemas are limited to RNN and LSTM opsets 1–6 and GRU opsets 3–6.
Upgrade to ONNX 1.22.0 or later. The fix validates that the first input has rank three before accessing its dimensions.
92e541266The original malformed model was also re-tested against the published ONNX 1.23.2 wheel and correctly raises InferenceError.
The vulnerable ONNX component processes the malicious model through a local library API or file boundary and is not itself bound to a network stack. Applications that expose ONNX model processing over a network should calculate an Environmental score with Modified Attack Vector set to Network.
onnx < 1.22.0Upgrade to a patched release:
onnx 1.22.0Connected by shared product, vendor, weakness, or advisory.
CVE-2026-115920Medium· 6.2ONNX: Out-of-bounds read in ConvTranspose shape inference via undersized weight rank
CVE-2026-115921Medium· 6.2ONNX: Null Pointer Dereference in ScanInferenceFunction Due to Missing Required Attribute
CVE-2026-115922Medium· 6.2ONNX: Denial of service via unchecked expansion of packed sub-byte tensors in ONNX Python APIs
CVE-2026-44512Medium· 5.5ONNX has Null Pointer Dereference in Upsample Version Converter Adapter (Zero Inputs)
CVE-2026-63632Low· 3.3Open Neural Network Exchange (ONNX) is an open standard for machine learning interoperability
CVE-2024-27318High· 7.5Onnx Directory Traversal vulnerability