README + API: add YouTube pitch link (Shorts)

Add GRPO walkthrough notebook to Hub; link from README and / API

Set HF_HOME and HF_DATASETS_CACHE for writable caches on Space

Set TORCHINDUCTOR cache dirs before torch import (HF Space no passwd uids)

Work around torch mega-cache double registration on Space imports

fix: durable outputs under /data, unbuffered training logs, relaxed deps

feat: Expose eval API endpoint

Post-training push to Hugging Face Hub: --hub-model-repo and pilot query params

write_test: explain repo vs container; add download_url and list_files

Add /outputs/write_test to create simpllll.csv and report hf_token env presence

Document HF token via Space secrets, ENV.example, and operational curl for pilot and push_to_hub

Add --bf16 flag and expose it via /train/pilot for bigger models

Allow model_name override on /train/pilot and raise num_generations cap

Fix reward collapse: floor exploration loss, bonus per-call and per-terminal

Live training visualization + aggressive reward shaping to prevent 'do nothing' collapse

Add compact training summary endpoint

Use Qwen3 model for TRL OpenEnv tool parsing

Add browser-accessible GRPO training endpoints

Add Space root health landing route

Deploy ReleaseOps Arena GPU Space