{"data":{"id":"6d139325-5929-4caf-8196-d496833092a2","title":"CVE-2026-73557: vLLM is an inference and serving engine for large language models. From 0.20.2rc0 until 0.26.0, safe_load_prompt_embeds ","summary":"vLLM (an AI inference and serving engine for large language models) versions 0.20.2rc0 to 0.26.0 have a vulnerability where concurrent requests to the chat API can bypass safety checks on prompt embeddings (pre-computed numerical representations of text input). This happens because the safety checking function uses a process-global state that can be exploited when multiple requests run simultaneously, potentially allowing invalid data to pass through even when safety features are enabled.","solution":"This issue is fixed in version 0.26.0. Users should update vLLM to version 0.26.0 or later.","labels":["security"],"sourceUrl":"https://nvd.nist.gov/vuln/detail/CVE-2026-73557","publishedAt":"2026-08-13T15:20:18.080Z","cveId":"CVE-2026-73557","cweIds":["CWE-362"],"cvssScore":null,"cvssSeverity":null,"severity":"high","attackType":[],"issueType":"vulnerability","affectedPackages":null,"affectedVendors":[],"affectedVendorsRaw":["vLLM"],"classifierModel":"claude-haiku-4-5-20251001","classifierPromptVersion":"v3","cvssVector":null,"attackVector":null,"attackComplexity":null,"privilegesRequired":null,"userInteraction":null,"exploitMaturity":"unknown","epssScore":0,"patchAvailable":null,"disclosureDate":"2026-08-13T15:20:18.080Z","capecIds":["CAPEC-26","CAPEC-29"],"crossRefCount":0,"attackSophistication":"moderate","impactType":["integrity","availability"],"aiComponentTargeted":"inference","llmSpecific":true,"classifierConfidence":0.85,"researchCategory":null,"atlasIds":null}}