#3842 LLM Resegment: Set OpenAI Model/Key in Settings.

The error "LLM Resegment: Set OpenAI Model/Key in Settings" indicates that the "LLM rephrase" feature is enabled, but the required OpenAI-compatible API configuration is missing.

Solution:

1. Go to the main menu: Tools/Options -> Advanced Options.
2. In the General Settings section, locate the option "LLM rephrase used AI channel".
3. Ensure this option is set to a valid AI channel that you have configured, such as OpenAI ChatGPT or DeepSeek.
4. You must have correctly entered the API URL and API Key for the selected channel in the Settings -> Translation Settings panel.
5. If you do not wish to use this feature, you can disable it. Go back to the main translation interface, click "Set more parameters...", and uncheck the "LLM rephrase" option.

Please refer to the relevant documentation:

• Optimizing segmentation effect, adjusting subtitle duration during speech recognition
• Using OpenAI's official ChatGPT API

Eu uso ollama

The error "LLM Resegment: Set OpenAI Model/Key in Settings" occurs because the "LLM rephrase" feature is enabled but configured to use an OpenAI-compatible channel (like OpenAI or DeepSeek) that lacks valid API credentials. You mentioned using Ollama, which is a local LLM.

Solution:

1. Configure the AI Channel for LLM Rephrase:

   • Go to the main menu: Tools/Options -> Advanced Options.
   • In the General Settings section, find the option "LLM rephrase used AI channel".
   • Change this setting to "Compatible AI/Local Model".

2. Configure the Local Model API:

   • Go to the main menu: Settings -> Translation Settings -> Local Large Model Settings.
   • In the API URL field, enter your Ollama server address, typically http://localhost:11434/v1.
   • In the Model field, enter the model name you are using with Ollama (e.g., qwen:7b, llama3.2:latest).
   • Leave the API Key field empty or fill it with a placeholder (like ollama), as Ollama usually does not require a key.

3. (Alternative) Disable LLM Rephrase:

   • If you do not need this feature, you can simply disable it. On the main translation interface, click "Set more parameters..." and uncheck the "LLM rephrase" option.

Please refer to the relevant documentation:

• Optimizing segmentation effect, adjusting subtitle duration during speech recognition
• Local Large Model and compatible AI interface usage