CVE-2026-61539

Xinference vulnerable to remote code execution via unsafe `eval()` in Llama3 tool-call parsing

Published 21 Aug 2026Modified 12 Oct 2026CWE 95

In the Trivy database since at least 24 Oct 2026, 07:11 UTCWhen it arrived before that build is not recorded.

Severity by source

GHSA10.0CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:H/A:H

Trivy's own label is CRITICAL. Rated critical by GHSA.

Description

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(), 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.

Affected packages and fixed versions

Source	Package	Fixed in
pip::GitHub Security Advisory pip	xinference	2.7.0

References

What Attestkeep does with this

Inside your cluster, without you opening a spreadsheet.

KnowsThe scanner refreshes its vulnerability database every six hours.
FindsEvery admitted image is rescanned every 24 hours by default (scan.rescanInterval). When Trivy rates this vulnerability critical for the image's package source, it appears in the image's findings by name. Findings without a fix are left out unless includeUnfixed is on.
Tells youA critical_finding event goes to the channels subscribed to it: e-mail, Slack, Teams, a webhook, Jira, GitLab or GitHub.
RefusesWith a policy that blocks critical findings, a pod with this image is refused at admission, a rollback to it included. Pods already running are not evicted.
RecordsEach decision is recorded with the policy that ran and its hash, then sealed under a signed checkpoint.

Try Attestkeep Read the docs

Vulnerability data compiled from the Trivy vulnerability database (trivy-db, Aqua Security, Apache License 2.0), which aggregates third-party advisories. CVE records are published by The MITRE Corporation under the CVE Terms of Use; CVE is a trademark of MITRE. This product uses data from the NVD API but is not endorsed or certified by the NVD. Severity, titles and advisory data from the GitHub Advisory Database (CC BY 4.0), Red Hat security data (CC BY 4.0) and the other sources linked on each entry. Entries may be summarised or re-formatted; the linked original is the authoritative record.