Add vllm docker info

Update config.json and hf_quant_config.json

Un-quantize MTP block (revert to native precision)

Drop FlashInfer MoE env vars from vLLM deploy command

Rename TensorRT Model Optimizer link to Model Optimizer (NVIDIA/Model-Optimizer)

Update nvidia-modelopt version to v0.44

Fix malformed Release Date to 05/27/2026

Remove stale Max OSL note from Evaluation section

Update README: replace eval table (GPQA/AA-LCR/τ²-Bench/SciCode/IFBench); switch runtime to SGLang and vLLM

Add files using upload-large-folder tool

Add files using upload-large-folder tool

initial commit