CVE-2026-61539

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.

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.

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, model_output: str
) -> List[Tuple[Optional[str], Optional[str], Optional[Dict[str, Any]]]]:
    try:
        data = eval(model_output, {}, {})
        return [(None, data["name"], data["parameters"])]
    except Exception:
        return [(model_output, None, None)]

The intended behavior was to convert a Python dictionary-like string generated by the model into a dictionary object. However, eval() executes the input as a Python expression, and eval(model_output, {}, {}) is not a security sandbox. If an attacker can influence the model output through prompt injection or direct chat input, the attacker can cause the model to return an expression such as:

__import__('os').system('touch /tmp/hacked')

When the expression reaches eval(), it is executed in the Xinference server process context. The harmless touch /tmp/hacked command can be replaced with other payloads, such as a reverse shell, malware download, sensitive file read, or lateral-movement payload.

Score

Severity: Critical

CVSS v3.1: 10.0

Vector: CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:H/A:H

Rationale:

• AV:N: the vulnerable API is remotely reachable over the network;
• AC:L: exploitation only requires a crafted chat-completion request and tool-call parameter;
• PR:N: the tested default configuration did not require authentication;
• UI:N: no user interaction is required;
• S:C: command execution can affect resources beyond the Xinference application boundary;
• C:H/I:H/A:H: remote code execution can fully compromise confidentiality, integrity, and availability.

Credit

This vulnerability was discovered by:

• XlabAI Team of Tencent Xuanwu Lab ([email protected])
• Atuin Automated Vulnerability Discovery Engine
• Guannan Wang ([email protected]), Zhanpeng Liu ([email protected]), Guancheng Li ([email protected])

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.

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.

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, model_output: str
) -> List[Tuple[Optional[str], Optional[str], Optional[Dict[str, Any]]]]:
    try:
        data = eval(model_output, {}, {})
        return [(None, data["name"], data["parameters"])]
    except Exception:
        return [(model_output, None, None)]

The intended behavior was to convert a Python dictionary-like string generated by the model into a dictionary object. However, eval() executes the input as a Python expression, and eval(model_output, {}, {}) is not a security sandbox. If an attacker can influence the model output through prompt injection or direct chat input, the attacker can cause the model to return an expression such as:

__import__('os').system('touch /tmp/hacked')

When the expression reaches eval(), it is executed in the Xinference server process context. The harmless touch /tmp/hacked command can be replaced with other payloads, such as a reverse shell, malware download, sensitive file read, or lateral-movement payload.

Score

Severity: Critical

CVSS v3.1: 10.0

Vector: CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:H/A:H

Rationale:

• AV:N: the vulnerable API is remotely reachable over the network;
• AC:L: exploitation only requires a crafted chat-completion request and tool-call parameter;
• PR:N: the tested default configuration did not require authentication;
• UI:N: no user interaction is required;
• S:C: command execution can affect resources beyond the Xinference application boundary;
• C:H/I:H/A:H: remote code execution can fully compromise confidentiality, integrity, and availability.

Credit

This vulnerability was discovered by:

• XlabAI Team of Tencent Xuanwu Lab ([email protected])
• Atuin Automated Vulnerability Discovery Engine
• Guannan Wang ([email protected]), Zhanpeng Liu ([email protected]), Guancheng Li ([email protected])

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, model_output: str
) -> List[Tuple[Optional[str], Optional[str], Optional[Dict[str, Any]]]]:
    try:
        data = eval(model_output, {}, {})
        return [(None, data["name"], data["parameters"])]
    except Exception:
        return [(model_output, None, None)]

The intended behavior was to convert a Python dictionary-like string generated by the model into a dictionary object. However, eval() executes the input as a Python expression, and eval(model_output, {}, {}) is not a security sandbox. If an attacker can influence the model output through prompt injection or direct chat input, the attacker can cause the model to return an expression such as:

__import__('os').system('touch /tmp/hacked')

When the expression reaches eval(), it is executed in the Xinference server process context. The harmless touch /tmp/hacked command can be replaced with other payloads, such as a reverse shell, malware download, sensitive file read, or lateral-movement payload.

Score

Severity: Critical

CVSS v3.1: 10.0

Vector: CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:H/A:H

Rationale:

• AV:N: the vulnerable API is remotely reachable over the network;
• AC:L: exploitation only requires a crafted chat-completion request and tool-call parameter;
• PR:N: the tested default configuration did not require authentication;
• UI:N: no user interaction is required;
• S:C: command execution can affect resources beyond the Xinference application boundary;
• C:H/I:H/A:H: remote code execution can fully compromise confidentiality, integrity, and availability.

Credit

This vulnerability was discovered by:

• XlabAI Team of Tencent Xuanwu Lab ([email protected])
• Atuin Automated Vulnerability Discovery Engine
• Guannan Wang ([email protected]), Zhanpeng Liu ([email protected]), Guancheng Li ([email protected])