{"answers":{"affected":"3 open source packages ship an affected version, including vllm.","exploited":"There is no known exploitation of CVE-2026-12491 in the wild and no public exploit code.","fix":"Upgrade vllm to 0.24.0 or later.","what":"CVE-2026-12491 is a medium-severity vulnerability in Red Hat AI Inference Server, published on 17 Jun 2026. An attacker can exploit it over the internet without logging in, with no action from the victim."},"fix_prompt":"Check this codebase for CVE-2026-12491 and fix it if it is affected.\n\nWhat it is: CVE-2026-12491 is a medium-severity vulnerability in Red Hat AI Inference Server, published on 17 Jun 2026. An attacker can exploit it over the internet without logging in, with no action from the victim.\nExploitation: There is no known exploitation of CVE-2026-12491 in the wild and no public exploit code.\n\nAffected packages and fixed versions:\n- vllm (PyPI): affected from 0.11.0 up to 0.23.0; no fix listed\n- vllm (PyPI): affected from 0.11.0; fixed in 0.24.0\n- vllm (PyPI): affected from 0.11.0; fixed in 0.24.0\n\nTask:\n1. Search the dependency manifests and lockfiles (for example package.json, package-lock.json, requirements.txt, poetry.lock, uv.lock, pom.xml, build.gradle, go.mod, Cargo.lock, Gemfile.lock) and any Dockerfiles for the affected packages, including transitive dependencies.\n2. If an affected version is in use, upgrade it to the fixed version for its release line with the project's package manager, so the lockfile is regenerated. Prefer the smallest upgrade that fixes it.\n3. If there is no fix, or the upgrade would break the build, do not guess: explain the risk and the safest workaround instead.\n4. Run the tests and summarise what changed and why.\n5. If nothing in the project is affected, say so and change nothing.\n\nReference: https://insights.sbomify.com/CVE-2026-12491/ (Markdown: https://insights.sbomify.com/CVE-2026-12491/index.md)","id":"CVE-2026-12491","markdown":"https://insights.sbomify.com/CVE-2026-12491/index.md","name":"CVE-2026-12491","record":{"aliases":["GHSA-8jr5-v98p-w75m","MINI-33jp-mfhh-vjp9","MINI-364f-6mjv-mhrh","MINI-g835-hm8q-2fpg","PYSEC-2026-3406"],"analysis":{"component_type":"library","fix_type":"patch","impacts":["data_modify"],"network_reachable":"yes","non_default_config":false,"product_category":"LLM inference library","requires_auth":"none","summary":"Improper image metadata handling in vLLM causes model input distortion, potentially leading to incorrect inference results.","user_interaction":"not_required","vuln_class":"other"},"analysis_generated_by":"qwen3.8-flash-next","assigner":"redhat","cpe":[],"cvss":{"score":4.8,"severity":"medium","source":"cna","vector":"CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:N/I:L/A:L","version":"3.1"},"cwes":[{"id":"CWE-115","name":"Misinterpretation of Input"}],"description":"A flaw was found in vLLM, an open-source library for large language model inference. This vulnerability arises from improper handling of image metadata, specifically EXIF orientation and PNG transparency (tRNS) data, during image processing. When images are converted to RGB, transparency information may be implicitly discarded or remapped, leading to unexpected rendering of transparent pixels and distortion of input content. This can result in the model misinterpreting image content, potentially affecting the integrity of processed data.","ecosystems":["PyPI"],"exploits":{"exploitdb":0,"first_seen":null,"github_poc":0,"max_poc_stars":null,"metasploit":0,"nuclei":0,"refs":[],"tier":"none_known"},"has_fix":true,"headline":"vLLM: image EXIF Rotation \u0026 PNG tRNS Transparency Not Normalized, Causing Mismatch Between Model Input and Expectations","id":"CVE-2026-12491","indexable":true,"kev":null,"kind":"cve","links":{"ecosystem":"PyPI","product":null,"products":[],"type":"other","year":"2026"},"n_packages":3,"n_references":2,"next":{"headline":"Happy Coders OTP Login for WooCommerce \u003c 2.8 - Unauthenticated Account Takeover via hcotp_auto_login_user","id":"CVE-2026-12492"},"packages":[{"ecosystem":"PyPI","fixed":[],"introduced":["0.11.0"],"last_affected":["0.23.0"],"package":"vllm","purl":"pkg:pypi/vllm","source":"PYSEC-2026-3406"},{"ecosystem":"PyPI","fixed":["0.24.0"],"introduced":["0.11.0"],"last_affected":[],"package":"vllm","purl":null,"source":"GHSA-8jr5-v98p-w75m"},{"ecosystem":"PyPI","fixed":["0.24.0"],"introduced":["0.11.0"],"last_affected":[],"package":"vllm","purl":"pkg:pypi/vllm","source":"GHSA-8jr5-v98p-w75m"}],"prev":{"headline":"Bypass of client certificate verification with transfer over TLS","id":"CVE-2026-12490"},"products":["Red Hat AI Inference Server","Red Hat Enterprise Linux AI (RHEL AI) 3","Red Hat OpenShift AI (RHOAI)","vLLM"],"published":"2026-06-17","related":[],"sources":[{"class":"attribution","licence":"CVE Terms of Use (MITRE)","name":"CVE List","prefix":"CVE"},{"class":"attribution","licence":"CC BY 4.0","name":"GitHub Advisory Database","prefix":"GHSA"},{"class":"attribution","licence":"CC BY 4.0","name":"PyPA Advisory Database","prefix":"PYSEC"},{"class":"attribution","licence":"CWE Terms of Use","name":"CWE (MITRE)"},{"class":"attribution","licence":"FIRST CVSS licence","name":"CVSS (FIRST)"}],"ssvc":{"automatable":"no","exploitation":"none","technical_impact":"partial"},"state":"PUBLISHED","updated":"2026-07-07","vendors":["Red Hat","vllm-project"]},"summary":"Improper image metadata handling in vLLM causes model input distortion, potentially leading to incorrect inference results.","url":"https://insights.sbomify.com/CVE-2026-12491/"}