{"id":34658967,"url":"https://github.com/deltartificial/g1","last_synced_at":"2026-04-17T05:02:04.015Z","repository":{"id":330303381,"uuid":"1122329564","full_name":"deltartificial/g1","owner":"deltartificial","description":"Simulate, train and deploy locomotion policies for Unitree G1 humanoid using Genesis on Mac M-series. 🦿","archived":false,"fork":false,"pushed_at":"2025-12-27T18:43:20.000Z","size":87,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2025-12-28T09:34:23.407Z","etag":null,"topics":["genesis","macos","python","reinforcement-learning","unitree"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/deltartificial.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2025-12-24T14:02:31.000Z","updated_at":"2025-12-27T18:42:22.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/deltartificial/g1","commit_stats":null,"previous_names":["deltartificial/g1"],"tags_count":12,"template":false,"template_full_name":null,"purl":"pkg:github/deltartificial/g1","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/deltartificial%2Fg1","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/deltartificial%2Fg1/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/deltartificial%2Fg1/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/deltartificial%2Fg1/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/deltartificial","download_url":"https://codeload.github.com/deltartificial/g1/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/deltartificial%2Fg1/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":31915900,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-04-16T18:22:33.417Z","status":"online","status_checked_at":"2026-04-17T02:00:06.879Z","response_time":62,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["genesis","macos","python","reinforcement-learning","unitree"],"created_at":"2025-12-24T18:38:27.793Z","updated_at":"2026-04-17T05:02:04.009Z","avatar_url":"https://github.com/deltartificial.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# G1\n\nSimulate, train and evaluate Unitree G1 humanoid robot using Genesis physics engine on Apple Silicon (Metal).\n\n## Setup\n\n```bash\npip install -r walk/requirements.txt\n```\n\n## Usage\n\n```bash\ncd walk\nmake train    # Train\nmake eval     # Evaluate\n```\n\n### Training\n\n```bash\nPYTHONUNBUFFERED=1 python3 -u g1_train.py -B 4096 --max_iter 1000\n```\n\n| Parameter | Description |\n|-----------|-------------|\n| `PYTHONUNBUFFERED=1` | Disable output buffering for real-time logs |\n| `-u` | Force unbuffered stdout/stderr |\n| `-B` | Number of parallel environments |\n| `--max_iter` | Maximum training iterations |\n| `--device` | Device: `mps` (default), `cuda`, `cpu` |\n| `--backend` | RL library: `auto`, `rsl_rl`, `sb3` |\n\n### Evaluation\n\n```bash\npython3 g1_eval.py --ckpt 700 --cmd_vel 0.8 0.0 0.0   # Fast forward walk\npython3 g1_eval.py --ckpt 700 --cmd_vel 0.3 0.2 0.0   # Lateral walk\n```\n\n| Parameter | Description |\n|-----------|-------------|\n| `--ckpt` | Checkpoint iteration to load |\n| `--cmd_vel vx vy vyaw` | Velocity command (m/s, m/s, rad/s) |\n\nJoint positions, velocities and link XYZ coordinates are logged automatically every 10 steps.\n\n`cmd_vel` controls robot movement:\n- `vx`: forward/backward speed\n- `vy`: lateral speed (left/right)\n- `vyaw`: rotation speed\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fdeltartificial%2Fg1","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fdeltartificial%2Fg1","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fdeltartificial%2Fg1/lists"}