EUVD-2026-41025
GHSA-vq94-fcwh-99v7
Severity by source
AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H
Remote unauthenticated low-complexity request to an exposed inference endpoint (AV:N/AC:L/PR:N/UI:N); decompression-bomb causes availability loss only, so C:N/I:N/A:H.
Primary rating from Vendor (nvidia).
CVSS VectorVendor: nvidia
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H
Lifecycle Timeline
2DescriptionCVE.org
NVIDIA Triton Inference Server for Linux contains a vulnerability where an attacker can cause improper handling of highly compressed data. A successful exploit of this vulnerability might lead to denial of service.
Articles & Coverage 1
AnalysisAI
Denial of service in NVIDIA Triton Inference Server for Linux allows remote unauthenticated attackers to exhaust server resources by submitting highly compressed (data-amplification / decompression-bomb) input that the server improperly handles during decompression. The flaw (CWE-409) affects the Linux distribution of Triton and carries a CVSS 7.5 (availability-only impact); there is no public exploit identified at time of analysis and it is not listed in CISA KEV. Impact is limited to availability - no confidentiality or integrity loss.
Technical ContextAI
NVIDIA Triton Inference Server is an open-source model-serving platform that exposes HTTP/REST and gRPC endpoints for running ML/AI models, commonly deployed behind GPU-backed inference infrastructure. The root cause is CWE-409 (Improper Handling of Highly Compressed Data, i.e. a data-amplification or 'decompression bomb' condition): the server accepts compressed request payloads and expands them without adequately bounding the decompressed size or the resources consumed, so a small malicious payload can inflate to consume disproportionate memory/CPU. The affected component is identified only generically by CPE cpe:2.3:a:nvidia:triton_inference_server:*:*:*:*:*:*:*:*, which does not pin a version range; the vulnerability is scoped to the Linux build of the product.
RemediationAI
Patch available per vendor advisory - consult the NVIDIA product-security bulletin at https://github.com/NVIDIA/product-security/tree/main/2026/5848 and upgrade to the fixed Triton Inference Server release listed there, as the provided data does not include an exact fixed version number to cite. Until patched, reduce exposure by not placing Triton endpoints on untrusted or internet-facing networks and restricting access to the HTTP/gRPC inference ports to trusted clients only (trade-off: breaks direct external client access); place a reverse proxy or API gateway in front that enforces request/body size limits and rejects or caps oversized or highly compressed payloads (trade-off: may reject legitimate large-but-benign requests and adds a hop); and apply resource limits such as container memory/CPU cgroup caps so a single decompression bomb degrades one instance rather than the host (trade-off: aggressive limits can cause premature OOM-kills of legitimate inference workloads). Verify NVD/NVIDIA references for the confirmed fixed build before deploying.
More in Triton Inference Server
View allAuthentication bypass in NVIDIA Triton Inference Server allows unauthenticated remote attackers to reach protected funct
Authentication bypass in NVIDIA Triton Inference Server allows remote unauthenticated attackers to circumvent access con
Integer overflow in the DALI backend of NVIDIA Triton Inference Server allows authenticated remote attackers to trigger
Out-of-bounds read in the DALI backend of NVIDIA Triton Inference Server allows authenticated remote attackers to trigge
Path traversal in NVIDIA Triton Inference Server for Linux (CWE-22) lets remote attackers manipulate file paths to reach
NVIDIA Triton Inference Server for Linux contains a vulnerability where a user can set the logging location to an arbitr
NVIDIA Triton Inference Server for Linux contains a vulnerability in the tracing API, where a user can corrupt system fi
NVIDIA Triton Inference Server for Linux contains a vulnerability in shared memory APIs, where a user can cause an impro
Denial of service in NVIDIA Triton Inference Server for Linux (versions through 26.03) allows a remote attacker to crash
Denial of service in NVIDIA Triton Inference Server for Linux stems from a stack-based buffer overflow (CWE-121) reachab
Denial of service in NVIDIA Triton Inference Server for Linux allows a remote, unauthenticated attacker to disrupt avail
Denial of service in NVIDIA Triton Inference Server on Linux allows remote unauthenticated attackers to exhaust host mem
Same technique Denial Of Service
View allShare
External POC / Exploit Code
Leaving vuln.today