What would you like to see?

Background

When configuring Azure OpenAI Service models, the model's Token Limit is intended to be selected from a predefined list of options.

Problem

Options corresponding to newer models, such as Reasoning models or GPT-4 Turbo, are missing from the selection list.
This prevents models with large context windows from utilizing their full capabilities.

Proposed Solution

• Add the missing options to the list defined in anything-llm/frontend/src/components/LLMSelection/AzureAiOptions/index.jsx.
• It may also be necessary to update the model context sizes defined in /anything-llm/server/utils/AiProviders/modelMap.js.

I plan to submit a Pull Request to address this issue later.