Live Exposure Defense: From CVE to Confirmed Exposure in 12 Hours – See more

Back
New CVE Detected

CVE-2026-61539 – Unauthenticated Remote Code Execution – Xinference (xorbitsai/inference) < 2.7.0

Be the first to know when new zero-days emerge:

Subscribe to Email Alerts
Subscribe to RSS Feed

Summary

CVE-2026-61539 is a critical eval-injection vulnerability (CWE-95) in Xinference, an open-source OpenAI-compatible inference server (xorbitsai/inference) used to serve open-source LLM, speech, and multimodal models. Versions 2.5.0 and earlier pass attacker-influenced Llama3 tool-call output directly into Python’s eval(), allowing an unauthenticated remote attacker to achieve arbitrary code execution on the server host via a crafted chat completion request. The flaw carries the maximum CVSS v3.1 base score of 10.0 (Critical).

Technical details

  • Root cause: The Llama3 tool-call parser (xinference/model/llm/tool_parsers/llama3_tool_parser.py) and a related helper in xinference/model/llm/utils.py evaluate model-generated text as Python code using eval(model_output, {}, {}). Restricting the globals/locals namespace does not prevent eval() from executing arbitrary Python expressions, so this is not an effective sandbox.
  • Trigger conditions: A request to the /v1/chat/completions OpenAI-compatible endpoint that includes a tools field. The request flows through xinference/api/restful_api.py and xinference/model/llm/transformers/core.py, into handle_chat_result_non_streaming() and _post_process_completion(), which call extract_tool_calls() or _eval_llama3_chat_arguments() — the functions that ultimately invoke eval() on model output.
  • Attack vector: Network, no authentication or user interaction required. An attacker crafts a prompt that manipulates the model into generating a malicious "tool call" argument string; when the server evaluates that string as Python, attacker-controlled code executes.
  • Impact: Full compromise of confidentiality, integrity, and availability — arbitrary command/code execution in the context of the Xinference server process, enabling data theft, model/data tampering, lateral movement, or denial of service.

Affected software

  • Vendor/Product: xorbitsai / inference (Xinference)
  • Affected versions: All versions up to and including 2.5.0 (versions prior to 2.7.0 that contain the vulnerable eval() code path)
  • Fixed version: 2.7.0

Severity

  • CVSS v3.1 Base Score: 10.0 (Critical)
  • Vector: CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:H/A:H

Mitigation and recommended actions

  • Immediate: Upgrade Xinference to version 2.7.0 or later, which removes the unsafe eval()-based tool-call parsing.
  • If immediate patching is not possible:
    • Restrict network exposure of the Xinference /v1/chat/completions API — do not expose it directly to the internet; place it behind authentication, an API gateway, or a network allow-list.
    • Disable or avoid using Llama3-based tool-calling functionality until patched.
    • Monitor Xinference server logs and host activity for anomalous process execution originating from the inference service.
    • Run the Xinference service with least-privilege OS permissions and in an isolated/sandboxed environment (e.g., container with restricted capabilities) to limit blast radius if exploited.

References

Are you exposed?

Get a free report of your organization’s exposure to this CVE and threat

  • Mapping of all Assets with this Technology
  • Identification of potentially exposed assets to this CVE
  • Confirmation of verified exploitable assets
Get Exposure Report

How IONIX’s External Exposure Management Platform Detects and Validates
Zero-Days to Shrink MTTR

1

Map your entire attack surface (continously)

IONIX uses multi-factor discovery methods, including DNS analysis, certificate mapping, metadata inspection, and more, to automatically map every internet-facing asset across your environment. This includes cloud instances, third-party platforms, shadow IT, and even forgotten infrastructure that traditional tools miss.

2

Monitor for new CVEs

Dozens of threat intel feeds using agentic technology are continuously analyzed to detect the appearance of proof-of-concept code, exploit kits, and indicators of active targeting. IONIX goes further by applying AI to proactively evaluate whether emerging vulnerabilities are likely to be exploited, even before PoCs go public.

3

Identify Potential External Exposures

Not all CVEs matter. IONIX filters vulnerabilities by asking attacker-centric questions: Can it be reached from the internet? Does it require authentication? Is it being exploited in the wild? This dramatically reduces noise and focuses teams on threats that can actually be weaponized.

4

Create Safe, Scalable Exploit Validations

IONIX transforms real-world PoCs into safe, non-intrusive test payloads that can be run in production environments without disruption. These simulations are precisely targeted to the systems that are vulnerable, ensuring rapid validation without unnecessary load.

5

Execute Exploit Validations

By combining context about software stack, versioning, exposure status, and reachability, IONIX ensures that only the right payloads are executed against the right assets, maximizing efficiency and minimizing risk.

6

Drive Fast and Actionable Remediation

Results are routed through integrations with ticketing, SOAR, and SIEM tools. Issues are written in plain language, bundled into remediation clusters, and prioritized based on asset criticality, exploitability, and blast radius. This shortens mean time to remediation (MTTR) and empowers teams to act with confidence.

Are you exposed?

Get a free report of your organization’s exposure to this CVE and threat

  • Mapping of all Assets with this Technology
  • Identification of potentially exposed assets to this CVE
  • Confirmation of verified exploitable assets
Get Exposure Report
Close

Subscribe to Threat Center RSS

Copy/paste the link below into your preferred RSS reader or follow these instructions to subscribe to Slack alerts.

Open RSS feedSlack setup instructions

Close

Get Real-Time CVE Alerts to Your Email

Be the first to know when new zero-days emerge