CanIEmbody
Matrix/reBot Arm B601-DM/π0.5

Does π0.5 run on reBot Arm B601-DM?

partial
T1 · upstream documentedlast verified 2026-10-25 · todayLeRobot

Scope: Training on reBot B601-DM data is documented by the vendor (Seeed wiki 'B601-DM with LeRobot'). The wiki's on-robot evaluation command (lerobot-record with --policy.path and no teleoperator) is rejected by the Seeed LeRobot fork the wiki installs, which directs policy deployment to lerobot-rollout, so on-robot inference is not validly documented. Public checkpoints trained on B601-DM data also exist.

Dispute this cell

Capabilities

CapabilityStateNotesEvidence
Train / fine-tuneworksev-seeed-wiki-b601-dm-lerobot, ev-hub-tinjyuu-pi05-rebot, ev-hub-qualia-pi05-rebot
Real-robot inferenceno verified evidence—
Simulation inferenceno verified evidence—
Teleop + recordno verified evidence—

Evidence 4 sources; the cell shows the highest tier

  1. T1 · upstream documentedvendor-docssource 2026-10-15verified 2026-10-25 (today)

    Seeed wiki 'B601-DM with LeRobot' (Training and Evaluation section). Pi0.5 tab: lerobot-train --policy.type=pi05 --policy.pretrained_path=lerobot/pi05_base on your reBot dataset, then evaluate with lerobot-record --robot.type=seeed_b601_dm_follower --policy.path=...

    “To use it in LeRobot: just set the policy type to `--policy.type=pi05`.”
    https://github.com/Seeed-Studio/wiki-documents/blob/86afe5634df5713aec98cf77c09a7e8d670feb1d/sites/en/docs/Robotics/Robot_Kits/reBot_Arm/B601_DM/reBot_Arm_B601_DM_Lerobot.md
  2. T1 · upstream documentedsource-codesource 2026-08-28verified 2026-10-25 (today)

    Seeed's LeRobot fork (main, pyproject version 0.6.2), which the wiki tells users to install: RecordConfig raises an error when no teleoperator is given and points policy deployment to lerobot-rollout.

    “For policy-based deployment, use lerobot-rollout instead.”
    https://github.com/Seeed-Projects/lerobot/blob/4aaff99be4a1d81568c08c8f0296b41b40c99ec4/src/lerobot/scripts/lerobot_record.py
  3. T3 · community reportedhub-checkpointverified 2026-10-24 (1d ago)

    config.json type=pi05; train_config.json dataset tinjyuu/rebot-prod1 whose meta/info.json robot_type=seeed_b601_dm_follower (Seeed plugin naming for the B601-DM, 43 episodes); pretrained from lerobot/pi05_base.

    https://huggingface.co/tinjyuu/pi05-rebot-prod1-peft-15k
  4. T3 · community reportedhub-checkpointverified 2026-10-24 (1d ago)

    config.json type=pi05; train_config.json dataset userresu2026/b601-dm-icehockey-v1 whose meta/info.json robot_type=seeed_b601_dm_follower (125 episodes).

    https://huggingface.co/qualia-robotics/pi05-b601-dm-icehockey-v1-72e6d068

π0.5 inference speed by hardware robot-independent · inference cost only

ComputeRuntimePrecisionCamsp50 / p95 msInferencePlatformSource
RTX 4090llama.cppfp162 @224x22451.7 / —19.34 Hz—
conditions

board: NVIDIA RTX 4090 · ggml_cuda_f16_f32_out: 0 (README preset for RTX 4090)

Inference cost only; not task success. Full VLM + action-expert pass (ViT + LLM + 10-step action expert), LIBERO inputs (agent-view + wrist at 224x224). From the EagleVLA-Edge README RTX 4090 table (P50), PR #6 configuration. The README's naive π0.5 baseline on this GPU: 162.9 ms (not filed as a separate run: the baseline precision is not stated).

T3 · community reported
RTX 4090llama.cppfp162 @224x22458.5 / —17.09 Hz—
conditions

board: NVIDIA RTX 4090 · ggml_cuda_f16_f32_out: 0 (README preset for RTX 4090)

Inference cost only; not task success. Full VLM + action-expert pass (ViT + LLM + 10-step action expert), LIBERO inputs (agent-view + wrist at 224x224). From the EagleVLA-Edge README RTX 4090 table (P50). The README's naive π0.5 baseline on this GPU: 162.9 ms (not filed as a separate run: the baseline precision is not stated).

T3 · community reported
Jetson AGX Thor 128GBllama.cppfp162 @224x224132.8 / —7.53 Hz—
conditions

board: Jetson Thor (paper/README: "Jetson Thor")

Inference cost only; not task success. Full VLM + action-expert pass (ViT + LLM + 10-step action expert), LIBERO inputs (agent-view + wrist at 224x224). Later configuration from the EagleVLA-Edge README (PR #6); the README notes an accuracy caveat for the two-switch GEMM preset on Thor. Naive PyTorch π0.5 on the same board per the paper: 457.9 ms (2.18 Hz) (not filed as a separate run: the baseline precision is not stated).

T3 · community reported
Jetson AGX Thor 128GBllama.cppfp162 @224x224202.8 / —4.93 Hz—
conditions

board: Jetson Thor (paper/README: "Jetson Thor")

Inference cost only; not task success. Full VLM + action-expert pass (ViT + LLM + 10-step action expert), LIBERO inputs (agent-view + wrist at 224x224). Paper's 'control frequency' (10.25 Hz) is 1/reaction time with VLM-skipping scheduling; not executed actions/s. Board inferred to be the 128 GB T5000 class from the paper's 130 W max-power entry for Thor (T4000 is 40–70 W). Naive PyTorch π0.5 on the same board per the paper: 457.9 ms (2.18 Hz) (not filed as a separate run: the baseline precision is not stated).

T3 · community reported
Jetson AGX Orin 64GBllama.cppfp162 @224x224271.1 / —3.69 Hzpower_mode 50W (paper label "Orin-50W")
conditions

board: Jetson Orin, 'Orin-50W' power mode (AGX Orin 64GB inferred, see ev-orin-board-inferred)

Inference cost only; not task success. Full VLM + action-expert pass (ViT + LLM + 10-step action expert), LIBERO inputs (agent-view + wrist at 224x224). Later configuration from the EagleVLA-Edge README (time cache + cuBLASLt GEMM tuning, PR #6); not in the paper. Naive PyTorch π0.5 on the same board per the paper: 1420.8 ms (0.70 Hz) (not filed as a separate run: the baseline precision is not stated).

T3 · community reported
Jetson AGX Orin 64GBllama.cppfp162 @224x224374.5 / —2.67 Hzpower_mode 50W (paper label "Orin-50W")
conditions

board: Jetson Orin, 'Orin-50W' power mode (AGX Orin 64GB inferred, see ev-orin-board-inferred)

Inference cost only; not task success. Full VLM + action-expert pass (ViT + LLM + 10-step action expert), LIBERO inputs (agent-view + wrist at 224x224). Paper's 'control frequency' (6.93 Hz) is 1/reaction time with confidence-based scheduling that skips the VLM; it is not executed actions/s and is not recorded as control_rate_hz. Naive PyTorch π0.5 on the same board per the paper: 1420.8 ms (0.70 Hz) (not filed as a separate run: the baseline precision is not stated).

T3 · community reported
Jetson Orin Nano 8GB (Super)TensorRTfp162 @640x480438.47 / 438.932.28 Hzjetpack 7.2.1 · power_mode MAXN_SUPER, clocks pinned with jetson_clocks (GPU 1020 MHz, CPU 1728 MHz) · tensorrt 10.16.2.10
conditions

board: Jetson Orin Nano Engineering Reference Developer Kit Super (8 GB) · l4t: R39.2.1 · os: Ubuntu 24.04.4 LTS · python: 3.12.3 · desktop: running

Inference cost only; not task success. 300 s run on synthetic 640x480 observations; 10 denoising steps; latency is runtime-only over a pre-materialized observation ring. Whole-board RAM in use 6.39 GB (unified memory, desktop running). Actions were gated against the stock policy (cosine >= 0.99999) before timing.

T3 · community reported
Jetson Orin Nano 8GB (Super)TensorRTfp163 @640x480643.39 / 644.921.55 Hzjetpack 7.2.1 · power_mode MAXN_SUPER, clocks pinned with jetson_clocks (GPU 1020 MHz, CPU 1728 MHz) · tensorrt 10.16.2.10
conditions

board: Jetson Orin Nano Engineering Reference Developer Kit Super (8 GB) · l4t: R39.2.1 · os: Ubuntu 24.04.4 LTS · python: 3.12.3 · desktop: running

Inference cost only; not task success. 300 s run on synthetic 640x480 observations; 10 denoising steps; latency is runtime-only over a pre-materialized observation ring. Whole-board RAM in use 6.18 GB (unified memory, desktop running). Actions were gated against the stock policy (cosine >= 0.99999) before timing.

T3 · community reported

Embed this cell

Badge for READMEs and model cards. It always shows the current state.

π0.5 on reBot Arm B601-DM: partial

[![π0.5 on reBot Arm B601-DM: partial](https://caniembody.pages.dev/badge/rebot-b601/pi05.svg)](https://caniembody.pages.dev/compat/rebot-b601/pi05)
<iframe src="https://caniembody.pages.dev/embed/rebot-b601/pi05" width="380" height="132" style="border:0" title="π0.5 on reBot Arm B601-DM | CanIEmbody"></iframe>

History

  1. 2026-10-25 fecccd1 Initial release: robot policy compatibility matrix · Danilo Znamerovszkij