# CVE-2026-61539: Xinference: Remote code execution via unsafe `eval()` in Llama3 tool-call parsing

> An unauthenticated remote attacker can execute arbitrary code on the Xinference server by crafting prompts that trigger unsafe eval() in Llama3 tool-call parsing.

- **Severity:** Critical 10.0 (`CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:H/A:H`)
- **Exploited in the wild:** Not known
- **Login needed:** Not needed
- **Attackable:** Over the internet
- **Victim action needed:** Not needed
- **Fix available:** Yes
- **Published:** 21 Aug 2026
- **Updated:** 24 Aug 2026
- **Also known as:** GHSA-x2rj-828p-hx9m, PYSEC-2026-3946

## What is CVE-2026-61539?

CVE-2026-61539 is a critical code injection vulnerability in xorbitsai inference, published on 21 Aug 2026. An attacker can exploit it over the internet without logging in, with no action from the victim.

### Official 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.

Weakness: CWE-95 (Improper Neutralization of Directives in Dynamically Evaluated Code)

## Which versions are affected by CVE-2026-61539?

3 open source packages ship an affected version, including xinference.


| Package | Ecosystem | Affected | Fixed in |
| --- | --- | --- | --- |
| `xinference` | PyPI | All versions | 2.7.0 |
| `xinference` | PyPI | All versions | 2.7.0 |
| `xinference` | PyPI | All versions | 2.7.0 |

## How do I fix CVE-2026-61539?

Upgrade xinference to 2.7.0 or later.

## Is CVE-2026-61539 being exploited?

There is no known exploitation of CVE-2026-61539 in the wild and no public exploit code.

## Fix prompt for AI coding agents

```text
Check this codebase for CVE-2026-61539 and fix it if it is affected.

What it is: CVE-2026-61539 is a critical code injection vulnerability in xorbitsai inference, published on 21 Aug 2026. An attacker can exploit it over the internet without logging in, with no action from the victim.
Exploitation: There is no known exploitation of CVE-2026-61539 in the wild and no public exploit code.

Affected packages and fixed versions:
- xinference (PyPI): affected all versions; fixed in 2.7.0
- xinference (PyPI): affected all versions; fixed in 2.7.0
- xinference (PyPI): affected all versions; fixed in 2.7.0

Task:
1. Search the dependency manifests and lockfiles (for example package.json, package-lock.json, requirements.txt, poetry.lock, uv.lock, pom.xml, build.gradle, go.mod, Cargo.lock, Gemfile.lock) and any Dockerfiles for the affected packages, including transitive dependencies.
2. If an affected version is in use, upgrade it to the fixed version for its release line with the project's package manager, so the lockfile is regenerated. Prefer the smallest upgrade that fixes it.
3. If there is no fix, or the upgrade would break the build, do not guess: explain the risk and the safest workaround instead.
4. Run the tests and summarise what changed and why.
5. If nothing in the project is affected, say so and change nothing.

Reference: https://insights.sbomify.com/CVE-2026-61539/ (Markdown: https://insights.sbomify.com/CVE-2026-61539/index.md)
```

## Sources

- CVE record: https://www.cve.org/CVERecord?id=CVE-2026-61539
- NVD: https://nvd.nist.gov/vuln/detail/CVE-2026-61539
- OSV: https://osv.dev/vulnerability/CVE-2026-61539

---
From sbomify Insights: https://insights.sbomify.com/CVE-2026-61539/
Summaries are generated and may contain mistakes; the official description above is the record.