docs: remove Gemma-4-12B-Agentic configuration from llama-swap.yaml for cleanup

docs: update hf reference in gemma-4-12b-agentic-fable5.ini for version consistency

docs: update spec-type and parameters in gemma-4-12b-agentic-fable5.ini for model configuration

docs: remove model-draft path from gemma-4-12b-agentic-fable5.ini for clarity

docs: update model-draft path in gemma-4-12b-agentic-fable5.ini for accuracy

docs: update model-draft path in gemma-4-12b-agentic-fable5.ini for version consistency

docs: update model-draft path in gemma-4-12b-agentic-fable5.ini for accuracy

docs: rename Gemma-4-12B-Agentic-Fable5 to Gemma-4-12B-Agentic for consistency

docs: update model-draft path and simplify spec-type in gemma-4-12b-agentic-fable5.ini

docs: update device configuration to support multi-GPU setup in gemma-4-12b-agentic-fable5.ini

docs: update device configuration and enhance spec-type settings in gemma-4-12b-agentic-fable5.ini

docs: update hf reference in gemma-4-12b-agentic-fable5.ini for consistency

docs: remove checkpoint settings and simplify spec-type in gemma-4-12b-agentic-fable5.ini

docs: add model-draft path to gemma-4-12b-agentic-fable5.ini for improved model loading

docs: update gemma-4-12b-it-qat.ini and gemma-4-12b-agentic-fable5.ini for enhanced specifications and add new agentic model configuration

docs: update gemma-4-12b-it-qat.ini to adjust np value and add chat-template-kwargs

docs: update gemma-4-12b-it-qat.ini to adjust np value and enable checkpoint settings

docs: update README and .env.example for clarity on configuration and service lifecycle

docs: enhance README with detailed preset file descriptions and usage examples

docs: update README to clarify repository purpose and contents

docs: add Hugging Face cache update script details to README

chore: remove unused PRESET variable and fix printf statement in stable-server script

chore: replace server and stable scripts with llama-server and stable-server scripts

Refactor and reorganize project structure

chore: update qwopus3.6-35b-mtp.ini with ctx-checkpoints and checkpoint-min-step settings

chore: comment out Qwen3.8-Flash-Next model configuration in llama-swap.yaml

chore: update configuration settings in qwen3.8-flash-next.ini for improved performance

chore: add load-mode and lazy-mode settings in qwen3.8-flash-next configuration

chore: enhance error handling and documentation in check-hf-updates script

chore: refactor check-hf-updates script for improved clarity and functionality

chore: improve documentation and refactor check-hf-updates script for better clarity and functionality

chore: enhance virtual environment setup and improve cache handling in check-hf-updates script

chore: refactor check-hf-updates script to add argument parsing and improve cache handling

chore: enhance virtual environment setup and cache directory resolution in check-hf-updates script

chore: add check-hf-updates script for managing Hugging Face model cache and virtual environment setup

chore: add Qwen3.8-Flash-Next model configuration and preset file

chore: update device configuration and refine spec settings in qwen3.8-27b-pi

chore: update device configuration and refine spec settings in qwen3.8-27b-pi

chore: update hf model reference in qwen3.8-27b-pi configuration

chore: simplify spec-type and comment out n-gram settings in qwen3.8-27b-pi configuration

chore: update spec-draft-n-max to 2 and add ctk, ctv settings in qwen3.8-27b-pi configuration

chore: update spec settings and add load-on-startup in qwen3.8-27b-pi configuration

chore: update context size to 65536 in qwen3.8-27b-pi configuration

chore: update context size for multiple models in llama-swap configuration

chore: update spec-draft-n-max to 4 and add context setting in configuration files

chore: standardize model names by removing suffixes in llama-swap configuration

chore: standardize section names by capitalizing preset headers in gemma-4-26b-a4b-it-qat-mtp and laya-bf16 configuration files

chore: standardize preset section names by removing suffixes in multiple configuration files

chore: increase context size from 65536 to 98304 in gemma-4-26b-a4b-it-qat-mtp preset

chore: adjust spec-draft-n-max to 3 and restore device settings in Qwen3.6-35B-MTP preset