“`html
A user reported an issue where audio input is not accepted by the LLaMA server when running with a Nemotron 3 nano Omni device. The user has tried using a recent build of llamacpp and redownloaded the Nemotron package, but the problem persists-audio input remains greyed out in the webUI, while image inputs are still working.
This issue seems to be specific to the compatibility between LLaMA’s server and the Nemotron 3 nano Omni device. The user has verified that other audio sources work correctly with different hardware setups, suggesting a potential problem within the llamacpp configuration for this particular device model.
- The reported issue highlights a lack of proper integration between LLama-server (the LLaMA server) and certain AI hardware like the Nemotron 3 nano Omni.
- It underscores the importance of thorough testing and compatibility checks when deploying AI models on new or custom hardware devices.
- Solving this issue may require updates to llamacpp’s configuration files for better support of the specific device model, which could be a collaborative effort between LLaMA developers and hardware manufacturers.
“`
Source Read original →
