new model

new model: Mythos-nano-heretic.Q4_K_M.gguf

new model

new model: bartowski/AGI-0_Art-Skynet-3B-GGUF

new model: mradermacher/Phi-4-mini-instruct-HERETIC-UNCENSORED-GGUF

new model: bartowski/Llama-3.2-3B-Instruct-uncensored-GGUF

fix model name: HauhauCS/Qwen3.5-4B-Uncensored-HauhauCS-Aggressive

new model: julien-c/Qwen3.5-4B-uncensored-GGUF

remove logger from openai endpoint

fix quotes syntaxerror

chunks logging

model response log

N_THREADS_BATCH type annotation error fix

speedup: N_THREAD_BATCH in config

tune n_thread back to 2 and n_thread_batch=2

Stop tracking VS Code and notebook checkpoints

try n_thread=4

model loading and warmup

llama model

Docker cmake update

docker update: CMAKE_ARGS=

API_KEY fix, model permission fixed, request locking fix

prod: implement thread-safe locking, bearer token auth, loguru, and eager model loading

multiple call handling and Docker model source change to config file

fix: resolve OOM 137 by limiting cmake build threads

readme update

feat: implement modular dual-endpoint FastAPI streaming service