feat: Add scripts for launching models with vLLM, including dynamic GPU allocation and logging functionality for better monitoring and management of model processes.

refactor: Simplify model configuration by replacing dynamic GPU mapping with a static dictionary, and enhance bot response function to include a seed value for reproducibility in responses.

chore: Update GPU mapping for model launches by commenting out unused configurations and adjust GPU memory utilization to 0.3. Add additional dependencies to requirements.txt for enhanced functionality.

refactor: Update model GPU mapping to a list format for improved readability and adjust GPU memory utilization settings for better performance during model launches.

fix: Update model launch message emoji for clarity and disable demo sharing option for improved privacy.

feat: Expand GPU mapping for model launching and enhance logging output with emojis for better visibility.

feat: Implement dynamic model launching with GPU mapping and update chatbot response handling to include thinking mode functionality.

feat: Integrate GPU support for model inference by adding the @spaces.GPU decorator to enhance performance during processing.

fix: Update chatbot configuration to specify message type for improved message handling in Model A and Model B outputs.

feat: Add new model options for inference and update requirements for OAuth support in Gradio.

feat: Introduce .gitignore and requirements.txt, rename bot_response to model_inference for clarity, and disable demo sharing option.

feat: Enable sharing option in demo launch for enhanced accessibility and collaboration.

refactor: Update model selection and local endpoint configuration for chatbot UI, adding support for multiple local endpoints and improving clarity in input handling.

refactor: Simplify chatbot configuration by removing redundant parameters and consolidating feedback logging functionality.

refactor: Clean up feedback logging code by removing unnecessary comments and ensuring consistent formatting.

refactor: Change feedback logging from CSV to JSONL format for improved data handling and consistency.

feat: Enhance feedback logging by switching from CSV to JSONL format and adding support for user message liking in chatbot UI.

feat: Implement feedback saving functionality with CSV logging and automatic syncing for user interactions in chatbots.

refactor: Update bot_response function to restore input box availability and clear content for improved user interaction.

refactor: Enhance bot_response function by improving error handling, refining message construction, and ensuring consistent UI updates during streaming.

refactor: Improve message handling in bot_response function by streamlining history updates and enhancing user interaction during streaming.

feat: Add support for local model endpoints and enhance bot_response function to handle local vs. remote inference clients.

refactor: Update UI labels and text for improved clarity and consistency across chatbot models

refactor: Enhance chatbot UI by updating comments, fixing clear logic, and ensuring message handling is consistent across models.

refactor: Simplify bot_response function by improving message handling and UI updates for user interactions.

Update

feat: Refactor chatbot functionality to support multiple models and improve user interaction with a new input system.

chore: Remove `google/gemma-7b-it` from the list of available models.

feat: Update Qwen model version in available models list

feat: Enable model selection via UI dropdown and dynamic InferenceClient initialization.

refactor: remove dynamic model selection UI and hardcode `openai/gpt-oss-20b` for the InferenceClient.

feat: Add model selection dropdown and dynamically use the chosen model in the inference client.

initial commit

Duplicate from gradio-templates/chatbot