vllm 0.0.15

Chart · Machine learning & AI · standard · v0.0.15

digest pinnedcosign signedSPDX SBOMSLSA provenanceamd64 · arm64

vLLM, the high-throughput LLM inference and serving engine with an OpenAI-compatible API. CPU build — installs vLLM's prebuilt CPU wheel (torch+cpu, no source compile, no CUDA) into a venv on a hardened Wolfi python-3.12 base, 0-CVE across ~140 Python packages. The user supplies the model; the chart runs it as a StatefulSet with a persistent model cache.

Version

0.0.16 · latest 0.0.15 0.0.14

The latest line lives at the base page; older lines have their own page so you can pin and verify exactly that version. Only the most recent releases are listed — an older chart pins an older image digest.

Released version

This is the 0.0.15 release of the vllm chart, published 2026-10-22. For the live security report and the currently deployed image digest, see the latest release.

Signed

cosign keyless

SBOM

SPDX, on image

Provenance

SLSA build

Architectures

amd64, arm64

Runs as

nonroot (uid 1001)

Root filesystem

read-only

Security report (Trivy)

D· 0/10017 fixable · rebuild clears them

Vulnerability detail

vllm 0.30.0 · 17 CVE

CVE	Severity	Package	Installed	Fixed in	Title
CVE-2026-114053	HIGH	vllm	0.30.0+cpu	0.31.0	vLLM: Request-controlled mm_processor_kwargs.code_revision allows remote code execution
CVE-2026-114055	HIGH	vllm	0.30.0+cpu	0.31.0	vllm: vLLM: Denial of Service via unvalidated multimodal processor arguments
CVE-2026-115912	HIGH	vllm	0.30.0+cpu	0.32.0	vllm: vllm: Arbitrary code execution via forged shared-memory handles
CVE-2026-115915	HIGH	vllm	0.30.0+cpu	0.32.0	vllm: vllm: Arbitrary code execution via unrestricted image format parsing
CVE-2026-114050	MEDIUM	vllm	0.30.0+cpu	0.31.0	vllm: vllm: Denial of Service via multimodal cache desynchronization
CVE-2026-115911	MEDIUM	vllm	0.30.0+cpu	0.32.0	vllm: vLLM: Denial of Service via unbounded chat template rendering
CVE-2026-115913	MEDIUM	vllm	0.30.0+cpu	0.32.0	vLLM: Phi-4 multimodal audio encoder materializes and retains full attention matrices: a single max-length audio OOM-crashes the vLLM engine
CVE-2026-115914	MEDIUM	vllm	0.30.0+cpu	0.32.0	vllm: vllm: Denial of service via crafted image input in OCR models
CVE-2026-115916	MEDIUM	vllm	0.30.0+cpu	0.32.0	vllm: vLLM: Denial of Service via out-of-bounds prompt token ID
CVE-2026-115918	MEDIUM	vllm	0.30.0+cpu	0.32.0	vllm: vLLM: Denial of service via untrusted chat template processing
CVE-2026-94624	MEDIUM	vllm	0.30.0+cpu	0.32.0	vllm: vLLM: Denial of Service via unbounded P2P KV offloading sessions
GHSA-c8m5-c952-6cm4	MEDIUM	rustls	0.23.45	0.23.46, 0.24.0-dev.2	Rustls: Session Debug formatting exposes TLS master secrets and resumption PSKs
CVE-2026-115917	LOW	vllm	0.30.0+cpu	0.32.0	vllm: vLLM: Information disclosure via permissive CORS configuration
GHSA-h2v8-394m-p7vv	LOW	rustls	0.23.45	0.23.46, 0.24.0-dev.2	Rustls: Brotli certificate decompression can allocate an attacker-controlled amount of memory up to 1 GiB before authentication.
GHSA-vp88-798v-h67f	LOW	rustls	0.23.45	0.23.46, 0.24.0-dev.2	rustls: ECH rejection can expose buffered application data to authenticated outer server before `ech_required` alert
GHSA-w44r-93j7-qv28	LOW	rustls	0.23.45	0.23.46, 0.24.0-dev.2	rustls: An unauthenticated ServerHello poisons the per-name key-exchange record, so later ClientHellos carry only a classical X25519 share
GHSA-xmj3-9gf7-h52w	LOW	setuptools	83.0.0	84.1.0	setuptools: Windows path escape via namespace_packages.txt during legacy wheel installation Description

Critical

High

Medium

Low

Unknown

Security report (Trivy) · image vllm 0.30.0

Install the chart

Deploy to Kubernetes with hardened defaults. The chart pins its image by signed digest, so you never track it yourself.

Install (pinned to 0.0.15)

helm install my-vllm oci://ghcr.io/quenchworks/charts/vllm --version 0.0.15

Chart version: 0.0.15
Chart license: Apache-2.0
App license: Apache-2.0
Service port: 8000
Signed: cosign (keyless)
Released: 2026-10-22

Verify the chart

cosign verify ghcr.io/quenchworks/charts/vllm:0.0.15 \
  --certificate-identity-regexp 'https://github.com/quenchworks/.+' \
  --certificate-oidc-issuer https://token.actions.githubusercontent.com

Transparency

The chart publishes its attestations on GitHub and the image it deploys carries its own on the same digest, publicly verifiable with the commands above. Both log to the Sigstore transparency log (Rekor), which cosign verify checks for you.

View the image View on ArtifactHub Chart source

Upstream project: https://docs.vllm.ai