{"slug":"openlair-verl-rl-training","source_name":"openlair/verl-rl-training","name":"Openlair/Verl Rl Training","description":"Provides guidance for training LLMs with reinforcement learning using verl (Volcano Engine RL). Use when implementing RLHF, GRPO, PPO, or other RL algorithms for LLM post-training at scale with flexible infrastructure backends.","version":1,"lift":{"pass_rate_delta_pts":50,"pass_rate_pct":77.3,"total_cases":22,"passed_cases":17,"tokens_delta_pct":86.1,"turns_delta_pct":0,"verdict":"mixed","benchmark_model":"gemini-3.6-flash","grading_method":"judged","completed_at":"2026-08-07T19:56:54.907066+00:00"},"skill_score":0.7727,"benchmark_models":[{"model":"gemini-3.6-flash","headline":true,"delta_pts":50,"with_pass_pct":77.3,"without_pass_pct":27.3,"tokens_delta_pct":86.1,"turns_delta_pct":0,"total_cases":22,"cases_aggregated":22,"verdict":"mixed","never_hurt":true,"completed_at":"2026-08-07T19:56:54.907066+00:00","run_id":"f288c915-2ff5-4a8f-8467-f37921807bc8","version_number":1,"is_latest_version":true,"gate":null}],"trust":{"skill_safety":"passed","safety_status":"clean","intent_verdict":"safe","content_status":"clean","indexable":true},"license":"MIT","install_count":0,"manifest_hash":"241d3b6694c00d58e99ea2623a929606f27cd0a65ed57e054dd80f8ebc2e2d3c","raw_url":"https://app.decimal.ai/s/openlair-verl-rl-training/SKILL.md","scorecard_url":"https://app.decimal.ai/skills/openlair-verl-rl-training"}