Four servable openpi checkpoints trained 2026-07-19 on
NoahWeiss/farm_uf850_home_full
(1153 episodes / 397k frames / 30 fps, UF850 arm, base + wrist RealSense):
Recipe: π0.5 base, PiSSA-initialized LoRA r64, masked loss + zero-padded noise
source on the 7 real action dims (of 32), M-draw 4, EMA 0.999, 15k steps,
batch 32. Each dir holds openpi-format params/ + assets/ (norm stats ride
along; optimizer state stripped).
1OPENPI_LORA_RANK=64 OPENPI_ZEROPAD=7 \
2uv run scripts/serve_policy.py --port 8000 policy:checkpoint \
3 --policy.config pi05_farm_clean \
4 --policy.dir <download>/farm_soup_s42/14999
Contract (must match — details + full arm runbook in
NoahWeiss123/farm model/FARM_CLEAN_ARM_RUNBOOK.md):
OPENPI_LORA_RANK=64 and
OPENPI_ZEROPAD=7 at serve; action_horizon=50;
outputs are ABSOLUTE joint targets (6 joints rad + gripper 0..1), 7 dims;
obs keys
observation/image (base cam),
observation/wrist_image,
observation/state (7,),
prompt = a trained task string verbatim;
224x224 resize-with-pad. Client:
model/eval_pi05.py with
--no-rtc.
The
pi05_farm_clean TrainConfig lives in the farm repo cluster patches
(clean-contract port of
pi05_farm_pissa with EMA + home_full).
