Cyber Resilience

CVE-2026-0897

DoS in Keras 3.0.0 – 3.13.0

Published
15 January 2026
Modified
15 July 2026
Patch / advisory
CVSS Score v4 7.1
Click a component to see what it means
Raw vectorCVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:P/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N/E:X/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:X
EPSS Score 0.0030 22th percentile
Risk Priority 55 floored blend · peak EPSS

Summary

CVE-2026-0897 is a high-severity Allocation of Resources Without Limits or Throttling (CWE-770) vulnerability in Keras Keras. Its CVSS base score is 7.1 (High).

Operationally, exploitation aligns with the MITRE ATT&CK technique Network Denial of Service (T1498); ranked at the 22th percentile by exploit likelihood (below the median); it is not currently listed in the CISA KEV catalog.

This vulnerability is AI-related — categorised as Deep Learning Frameworks; in the Data-Related Vulnerabilities risk domain.

The strongest mitigations our analysis identified map to AC-10 (Concurrent Session Control) and SC-6 (Resource Availability) — see the control section below for these in your framework.

Deeper analysis AI-assisted summary

Synthesised by an AI model from the NVD description and linked references — a reading aid, not an authoritative source.

CVE-2026-0897 is a vulnerability classified under CWE-770 (Allocation of Resources Without Limits or Throttling) in the HDF5 weight loading component of Google Keras versions 3.0.0 through 3.13.0, affecting all platforms. The issue enables a remote attacker to trigger a Denial of Service (DoS) condition through memory exhaustion and a crash of the Python interpreter. This occurs via a crafted .keras archive that includes a valid model.weights.h5 file where the dataset declares an extremely large shape.

Any remote attacker can exploit this vulnerability over the network with low complexity, no privileges, and no user interaction required, as reflected in its CVSS v3.1 base score of 7.5 (AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H). Exploitation involves tricking a target into loading the malicious .keras archive, resulting in unbounded memory allocation during HDF5 dataset processing, which exhausts system resources and crashes the Python interpreter.

A pull request addressing the issue is available at https://github.com/keras-team/keras/pull/21880, providing details on the patch for mitigation in affected Keras versions.

EU & UK References

Vulnerability Data

Allocation of Resources Without Limits or Throttling in the HDF5 weight loading component in Google Keras 3.0.0 through 3.13.0 on all platforms allows a remote attacker to cause a Denial of Service (DoS) through memory exhaustion and a crash of…

more

the Python interpreter via a crafted .keras archive containing a valid model.weights.h5 file whose dataset declares an extremely large shape.

CWE(s)

AI Security AnalysisAI

AI Category
Deep Learning Frameworks
Risk Domain
Data-Related Vulnerabilities
OWASP Top 10 for LLMs 2025
None mapped
Classification Reason
Matched keywords: keras

Related Threats

MITRE ATT&CK Enterprise Techniques

T1498 Network Denial of Service Impact
Adversaries may perform Network Denial of Service (DoS) attacks to degrade or block the availability of targeted resources to users.
T1499 Endpoint Denial of Service Impact
Adversaries may perform Endpoint Denial of Service (DoS) attacks to degrade or block the availability of services to users.
T1499.003 Application Exhaustion Flood Impact
Adversaries may target resource intensive features of applications to cause a denial of service (DoS), denying availability to those applications.
T1498.001 Direct Network Flood Impact
Adversaries may attempt to cause a denial of service (DoS) by directly sending a high-volume of network traffic to a target.
T1498.002 Reflection Amplification Impact
Adversaries may attempt to cause a denial of service (DoS) by reflecting a high-volume of network traffic to a target.
T1499.001 OS Exhaustion Flood Impact
Adversaries may launch a denial of service (DoS) attack targeting an endpoint's operating system (OS).
Derived from this CVE’s CWE(s) via the direct CWE→ATT&CK cross-walk.

CVEs Like This One

CVE-2026-1669Same product: Keras Keras
CVE-2025-9905Same product: Keras Keras
CVE-2025-8747Same product: Keras Keras
CVE-2026-11816Same product: Keras Keras
CVE-2024-55459Same product: Keras Keras
CVE-2025-9906Same product: Keras Keras
CVE-2026-12481Same product: Keras Keras
CVE-2024-3660Same product: Keras Keras
CVE-2025-1550Same product: Keras Keras
CVE-2026-12482Same product: Keras Keras

Affected Assets

keras
keras
3.0.0 — 3.13.0

Mitigating Controls

Control response

Prevent
Stop it (NIST 800-53)

Detect
Catch it (NIST detect / respond)

Harden
Shrink the surface (DISA STIG)
  • 5 hardening rules · 3 OS baselines
Validate
Prove the fix (OWASP ASVS)
  • V15.4.4

Mitigating Controls (NIST 800-53 r5) AI

Directly enforces a hard limit on concurrent sessions, structurally preventing unbounded resource allocation.

Requires explicit allocation of resources by priority or quota, directly stopping unlimited allocation.

Imposes a limit on consecutive invalid attempts, preventing one specific class of unbounded resource consumption.

Reduces impact of resulting DoS events without preventing the underlying lack of allocation limits.

Mitigating Controls (NIST CSF 2.0) AI

Derived directly from the weakness types (CWEs) cited in the NVD entry via our AI-authored CWE→CSF cross-walk (authority under review) — links open the control.

PR.IR-04 mostly match
degrades

Monitoring capacity and taking action to maintain availability directly reduces unchecked resource allocation.

Mitigating Controls (ISO/IEC 27001:2022 Annex A) AI

Derived directly from the weakness types (CWEs) cited in the NVD entry via our AI-authored CWE→ISO cross-walk (authority under review) — links open the control.

finds

Baseline comparison of CPU, memory and bandwidth usage helps surface uncontrolled resource allocations before they cause service degradation.

prevents

Capacity projections and elasticity measures ensure that allocation requests are bounded and can be throttled, reducing the window in which an attacker can force unbounded resource reservations.

mitigates

Defining retention periods and deletion schedules for backup copies prevents indefinite accumulation of data on storage media without corresponding resource-management controls.

mitigates

Architectural redundancy and automatic failover limit the impact of an attacker who forces excessive allocations, because spare capacity can absorb the load until the primary instance recovers.

none

Documented incident response procedures that include activation of continuity plans and controlled recovery help ensure that resource consumption triggered by an incident is bounded and managed rather than left unbounded.

mitigates

Mandating tested continuity procedures that preserve or replace resource-limiting controls prevents an attacker from exploiting the absence of throttling mechanisms during an outage.

Hardening callouts derived

Configuration rules from DISA STIG baselines that bear on weaknesses of the type cited by this CVE. Each rule is shown with the relationship its mapping actually records, against the CWE it was authored against. Derived via CVE→CWE over `controls_xwalks` (authoritative rows only; rows rated `none` are excluded).

Oracle Linux 8 (2 rules)
  • V-248552 OL 8 must be configured so that all network connections associated with SSH traffic terminate after becoming unresponsive. prevents CWE-770
  • V-248553 OL 8 must be configured so that all network connections associated with SSH traffic are terminated after 10 minutes of becoming unresponsive. prevents CWE-770
Oracle Linux 9 (2 rules)
  • V-271710 OL 9 must be configured so that all network connections associated with SSH traffic are terminated after 10 minutes of becoming unresponsive. prevents CWE-770
  • V-271709 OL 9 must be configured so that all network connections associated with SSH traffic terminate after becoming unresponsive. prevents CWE-770
RHEL 8 (1 rule)
  • V-230244 RHEL 8 must be configured so that all network connections associated with SSH traffic terminate after becoming unresponsive. prevents CWE-770

References