Vulnerability in xinference (CVE-2026-61539)
Summary
Vulnerability in xinference (CVE-2026-61539). Successful exploitation can lead to full system takeover. Exploitable via tools. Mitigation: upgrade to 2.7.0 or later.
Original description
The description as published by the source.
Summary
Xinference used Python's unsafe eval() function when parsing Llama3 tool-call output generated by a large language model. Because the model output can be influenced by attacker-controlled prompts sent to the chat completion API, a remote attacker can craft prompts that cause the model to return a Python expression. Xinference then evaluates that expression on the server while post-processing the tool-call result. In the tested default deployment, authentication was not enabled, so the vulnerability was exploitable by an unauthenticated remote attacker through the /v1/chat/completions endpoint.
Details
Users can interact with deployed models through Xinference's OpenAI-compatible /v1/chat/completions API. The request entry point is implemented in xinference/api/restful_api.py; non-streaming requests call the model instance's chat() method and return the inference result.
When the Transformers backend is used, inference results flow through the batching logic in xinference/model/llm/transformers/core.py. Non-streaming chat results are handled by handle_chat_result_non_streaming(). If the request contains a tools field, Xinference calls _post_process_completion() to parse tool-call output from the model response.
The Llama3 tool-call parser is implemented in xinference/model/llm/tool_parsers/llama3_tool_parser.py. In affected versions, extract_tool_calls() parsed model output with eval():
def extract_tool_calls(
self, m…
Source: GitHub Security Advisories (Global) NVD GitHub Advisory
Auto-generated summary
Assembled by SecPulse from the source data (CVSS, CWE, affected versions) using templates — no AI involved. Check the original description and references below for exact details.
A vulnerability tracked as CVE-2026-61539 has been found in xinference.
Attackers can target a specific entry point like tools over the network to misuse the product.
Successful exploitation can lead to full system takeover. CVSS score: 10.0/10.
What to do: upgrade xinference to 2.7.0 or later.
If unsure, ask your IT team or search "xinference CVE-2026-61539" on the vendor's site.
CVE-2026-61539 (xinference) — CWE-95 / CVSS v3 10.0
Attack vector: remote (network-reachable) / unauthenticated / no user interaction
Attack surface: tools
Patched: 2.7.0 — apply immediately
Plan: 1) Audit SBOM/dependencies, 2) Stage→prod upgrade, 3) Add WAF/proxy monitoring on affected endpoints, 4) Hunt IOCs in logs.
Refs: see the GHSA / vendor advisory / patched release linked on this page.
A vulnerability (CWE-95) exists in xinference. Attackers reach the vulnerable code path via tools without authentication.
Check the vendor advisory for the affected versions of xinference. Attack surface: tools.
Severity: Critical (CVSS 10.0/10). Successful exploitation can lead to full system takeover.
Update to 2.7.0.
Until the patch is applied: disable the affected feature, apply WAF rules, or restrict access via network ACLs.
Search webserver/proxy logs for unusual request patterns matching this CVE's known IOCs. Run grep -r 'xinference' . against your dependency files (package-lock.json, requirements.txt, go.sum) to find affected services.
Response actions
Concrete steps and command examples for SOC/SRE teams to execute in order
-
1Identify exposure identify
grep -r 'xinference' . | grep -v node_modulesリポジトリと本番環境の依存ファイル (package-lock.json / requirements.txt / go.sum / Gemfile.lock 等) で
xinferenceを grep し、稼働しているサービス・バージョンを把握する。 -
2Apply patch patch
Upgrade xinference to 2.7.0ステージング環境で 2.7.0 に上げて回帰テスト → 本番反映。回帰テストはアプリの主要ハッピーパスを含めること。
-
3Post-deployment verification verify
Confirm patched version is live in productionパッチ適用後、ステージングで PoC または同等の悪用パターンを再現して脆弱性が閉じたことを確認。本番ではアクセスログ・WAF でアラート再発が無いか継続監視。