Raw vector
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:H/A:HCVSS and EPSS are reproduced from their sources (NVD, FIRST EPSS). Risk Priority is our own derived reading, not an NVD score.
Summary
CVE-2026-61539 is a critical-severity Eval Injection (CWE-95) vulnerability. Its CVSS base score is 10.0 (Critical).
Operationally, exploitation aligns with the MITRE ATT&CK technique Command and Scripting Interpreter (T1059); ranked in the top 32% of CVEs by exploit likelihood; it is not currently listed in the CISA KEV catalog.
This vulnerability is AI-related — categorised as LLM Application Platforms.
The strongest mitigations our analysis identified map to SA-11 (Developer Testing and Evaluation) and SI-10 (Information Input Validation) — see the control section below for these in your framework.
OWASP Top 10 for Web (2025)
EU & UK References
- 🇪🇺 ENISA EUVD: EUVD-2026-64146
Vulnerability Data
Xinference is an inference API for running open-source, speech, and multimodal models. In 2.5.0 and earlier, Xinference passes attacker-influenced Llama3 tool-call output to eval() in xinference/model/llm/tool_parsers/llama3_tool_parser.py and xinference/model/llm/utils.py. Requests to /v1/chat/completions with a tools field flow through xinference/api/restful_api.py, xinference/model/llm/transformers/core.py, handle_chat_result_non_streaming(),…
more
and _post_process_completion() before extract_tool_calls() or _eval_llama3_chat_arguments() evaluates the model-generated Python expression. An unauthenticated remote attacker can influence that output through a crafted prompt and execute commands in the Xinference server process context. This issue is fixed in version 2.7.0.
- CWE(s)
AI Security AnalysisAI
- AI Category
- LLM Application Platforms
- Risk Domain
- N/A
- OWASP Top 10 for LLMs 2025
- None mapped
- AI-specific weaknesses CR
- CWE-1426 — Model output reaches eval() without validation; prompt injection is delivery vector.
Mapped by Cyber Resilience · not in NVD. Poisoning and extraction cases are routed to MITRE ATLAS instead of a synthetic CWE.- Classification Reason
- Matched keywords: llm, llm, llm, transformers
Related Threats
MITRE ATT&CK Enterprise Techniques
CVEs Like This One
Affected Assets
Mitigating Controls
Control response
—
—
—
V1.3.2
Mitigating Controls (NIST 800-53 r5) AI
Developer testing and code analysis can discover eval-injection flaws but does not stop their introduction.
Input validation explicitly requires neutralizing untrusted data before it reaches dynamic evaluation constructs such as eval.
Secure-development standards and tools can mandate safe coding patterns that avoid unsafe dynamic evaluation.
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.
Secure SDLC practices directly require input neutralization and avoidance of unsafe dynamic evaluation.
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.