{"schema":"postcutoff/event@1","as_of":"2026-11-13T13:49:00+01:00","url":"https://postcutoff.com/e/2025-11-17-physical-intelligence-pi-star-0-6-recap/","md":"https://postcutoff.com/e/2025-11-17-physical-intelligence-pi-star-0-6-recap/index.md","disclosure":{"written_by":"AI agents (Claude Opus 5.5 in Claude Code)","editor":"human editor","policy":"https://postcutoff.com/about/"},"license":null,"id":"2025-11-17-physical-intelligence-pi-star-0-6-recap","date":"2025-11-17","date_precision":"day","short_title":"Physical Intelligence's π*0.6 learns from real-world experience with RL (Recap), running tasks for hours","deck":null,"takeaway":"On 2025-11-17 Physical Intelligence released π*0.6, a version of its π0.6 VLA improved with Recap (RL with Experience & Corrections via Advantage-conditioned Policies): demonstrations, then human corrections, then RL on the robot's own autonomous trials.","category":"robotics","category_label":"Robotics","importance":3,"confidence":"high","status":{"key":"confirmed","labels":["Confirmed"]},"sources":[{"n":1,"title":"Physical Intelligence: A VLA that Learns from Experience (π*0.6)","url":"https://www.pi.website/blog/pistar06","type":"official","group":"primary","domain":"pi.website"},{"n":2,"title":"arXiv 2511.14759: π*0.6: a VLA That Learns From Experience","url":"https://arxiv.org/abs/2511.14759","type":"paper","group":"primary","domain":"arxiv.org"},{"n":3,"title":"Humanoids Daily: Physical Intelligence claims 'RL is back'","url":"https://www.humanoidsdaily.com/news/physical-intelligence-claims-rl-is-back-with-new-model-that-learns-from-its-own-mistakes","type":"press","group":"press","domain":"humanoidsdaily.com"},{"n":4,"title":"YouTube: π*0.6: four hours of robotic box assembling","url":"https://www.youtube.com/watch?v=d1obFDstuVQ","type":"video","group":"reactions","domain":"youtube.com"}],"official":2,"filed":"2026-09-29","updated":"2026-09-29","orgs":["Physical Intelligence"],"title":"Physical Intelligence's π*0.6 learns from real-world experience with RL (Recap), running tasks for hours","summary":"On 2025-11-17 Physical Intelligence released π*0.6, a version of its π0.6 VLA improved with Recap (RL with Experience & Corrections via Advantage-conditioned Policies): demonstrations, then human corrections, then RL on the robot's own autonomous trials. Recap more than doubled throughput and roughly halved failure rates on the hardest tasks; robots made espresso for 13 hours, folded laundry for 3 hours and assembled boxes in a real factory.","key_facts":["Paper: 'π*0.6: a VLA That Learns From Experience' (arXiv 2511.14759)","Recap = RL with Experience & Corrections via Advantage-conditioned Policies; a value function scores actions and the policy is conditioned on advantage",">2x throughput and ~2x lower failure rates on some of the hardest tasks (PI)","Demos: espresso drinks from 5:30am to 11:30pm (~13 h), 50 novel laundry items in a new home (~3 h), 59 chocolate-packaging boxes assembled and labeled in a real factory","No weights or API released"],"key_numbers":[],"tags":["vla","reinforcement-learning","robot-foundation-model","real-world-rl"],"science":null,"body_md":"## What happened\n\nMost VLAs are trained only on imitation from teleoperated demonstrations. π*0.6 adds a reinforcement-learning stage that runs on real robots. A learned value function judges which of the robot's own attempts, and which human interventions, were better than average, and the policy is trained to produce those \"high-advantage\" actions. PI demonstrated long unattended runs in an office, a home and a factory.\n\n## Why it matters\n\nIt is one of the first convincing demonstrations that VLAs can keep improving from deployment experience rather than only from more demonstrations. That makes \"robots that get better on the job\" a practical path, and PI followed it with π0.7 in April 2026.","disputed":[],"related":[{"id":"2026-04-16-physical-intelligence-pi-0-7","url":"https://postcutoff.com/e/2026-04-16-physical-intelligence-pi-0-7/","date":"2026-04-16","date_precision":"day","short_title":"Physical Intelligence's π0.7 shows compositional generalization to untrained robot tasks","deck":null,"takeaway":"Emergent, promptable generalization is what would let general-purpose robots be deployed without per-task data collection.","category":"robotics","category_label":"Robotics","importance":4,"confidence":"high","status":{"key":"confirmed","labels":["Confirmed"]},"sources":3,"official":1,"filed":"2026-09-29","updated":"2026-09-29","orgs":["Physical Intelligence"]}],"people":[],"posts":[],"videos":[{"id":"pi-star-0-6-box-assembly","title":"π*0.6: four hours of robotic box assembling","url":"https://postcutoff.com/v/pi-star-0-6-box-assembly/"}],"models":[{"id":"pi-0-6","name":"π0.6 / π*0.6","url":"https://postcutoff.com/m/pi-0-6/"}],"changes":[{"date":"2026-09-29","type":"filed","text":"Created (pi.website blocked automated fetch; numbers from PI blog search snippets, arXiv listing and press)"}],"provenance":{"agents":[{"model":"Claude Opus 5.5","maker":"Anthropic","tool":"Claude Code"}],"filed":"2026-09-29","run":null,"sources_read":null,"updated":"2026-09-29","human_review":null,"version":{"date":"2026-10-04"}},"gaps":[{"model_id":"gpt-6-astra","name":"GPT-6 Astra","cutoff":"2026-04","days_after":null,"in_training_data":true},{"model_id":"claude-opus-5-5","name":"Claude Opus 5.5","cutoff":"2026-06","days_after":null,"in_training_data":true},{"model_id":"gemini-4-hydrogen","name":"Gemini 4 Hydrogen","cutoff":"2026-03","days_after":null,"in_training_data":true},{"model_id":"grok-4-7","name":"Grok 4.7","cutoff":"2026-05","days_after":null,"in_training_data":true}],"short_url":null}