CVE-2026-112080: LlamaChinese Llama-Chinese Adapter Config chat_gradio_no_merge.py get_class_from_dynamic_module code injection
A vulnerability was found in LlamaChinese Llama-Chinese up to 6fa0fffb0dd82fe3cfaa1449ee54a5806d26ae9b. The impacted element is the function getclassfromdynamicmodule of the file examples/chatgradionomerge.py of the component Adapter Config. The manipulation of the argument basemodelnameorpath results in code injection. The attack may be performed from remote. The exploit has been made public and could be used. This product utilizes a rolling release system for continuous delivery, and as such, version information for affected or updated releases is not disclosed. The project was informed of the problem early through an issue report but has not responded yet.
Affected Software
Event History
Frequently Asked Questions
What must an attacker be able to control to exploit this issue?
An attacker needs to manipulate the base_model_name_or_path argument used by get_class_from_dynamic_module in examples/chat_gradio_no_merge.py. The issue can be exploited remotely and does not require privileges, although user interaction is required according to the supplied severity vector.
Is there a known fixed release?
No fixed version is disclosed. The project uses a rolling release model, so affected and updated release version information is not available, and the project had not responded to the reported issue.
What should teams do if they cannot confirm a patch is available?
Do not allow untrusted users or remote inputs to control base_model_name_or_path in the affected example script. Restrict access to the deployment and use only trusted, locally controlled model paths or adapter configuration values.
How urgent is remediation?
The issue is rated medium severity with a 6.3 CVSS score, but a public exploit is available. Prioritize systems where the affected script is remotely reachable or where untrusted parties can influence adapter or model configuration.