Does π0.5 run on I2RT YAM?
Capabilities
| Capability | State | Notes | Evidence |
|---|---|---|---|
| Train / fine-tune | works | ev-hub-asethi-pi05, ev-hub-nuffnuff-pi05 | |
| Real-robot inference | works with caveats | ev-molmoact2-paper-c3 | |
| Simulation inference | no verified evidence | — | |
| Teleop + record | no verified evidence | — |
Caveats
- Community-reported only; not documented upstream or maintainer-tested.
- Real-robot results come from a π0.5-DROID baseline in the MolmoAct2 paper (average 32.2% over 8 bimanual YAM tasks), which does not name the codebase used; the LeRobot checkpoints' cards do not report hardware rollouts.
- LeRobot has no built-in YAM robot type; the checkpoints use bi_yam_follower / yam_follower_bimanual types from third-party plugins.
Evidence 3 sources; the cell shows the highest tier
- T3 · community reportedhub-checkpointsource 2026-08-06verified 2026-10-24 (1d ago)
config.json type=pi05 (fine-tuned from lerobot/pi05_base on a LeRobot fork whose pi05 code matches upstream main); trained on brandonyang/yam-vive-teleop whose meta/info.json robot_type=bi_yam_follower. Card: hardware rollouts pending.
https://huggingface.co/ASethi04/pi05-BimanualYAM-oranges - T3 · community reportedhub-checkpointsource 2026-05-29verified 2026-10-24 (1d ago)
config.json type=pi05; trained on nuffnuff/yam-master-v1 whose meta/info.json robot_type=yam_follower_bimanual.
https://huggingface.co/nuffnuff/pi05-yam-finetuned - T3 · community reportedpapersource 2026-05-04verified 2026-10-24 (1d ago)
MolmoAct2 paper appendix C.3: π0.5-DROID baseline fine-tuned per task and evaluated on real bimanual YAM, 8 tasks x 50 trials (Table 19 average 32.2%).
“to ensure a fair comparison, we applied the identical training protocol to each of the four baseline policies. All policies were trained to convergence, and the resulting checkpoints were used for evaluation.”
https://arxiv.org/abs/2605.02881
π0.5 inference speed by hardware robot-independent · inference cost only
| Compute | Runtime | Precision | Cams | p50 / p95 ms | Inference | Platform | Source |
|---|---|---|---|---|---|---|---|
| RTX 4090 | llama.cpp | fp16 | 2 @224x224 | 51.7 / — | 19.34 Hz | —conditionsboard: 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 4090 | llama.cpp | fp16 | 2 @224x224 | 58.5 / — | 17.09 Hz | —conditionsboard: 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 128GB | llama.cpp | fp16 | 2 @224x224 | 132.8 / — | 7.53 Hz | —conditionsboard: 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 128GB | llama.cpp | fp16 | 2 @224x224 | 202.8 / — | 4.93 Hz | —conditionsboard: 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 64GB | llama.cpp | fp16 | 2 @224x224 | 271.1 / — | 3.69 Hz | power_mode 50W (paper label "Orin-50W")conditionsboard: 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 64GB | llama.cpp | fp16 | 2 @224x224 | 374.5 / — | 2.67 Hz | power_mode 50W (paper label "Orin-50W")conditionsboard: 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) | TensorRT | fp16 | 2 @640x480 | 438.47 / 438.93 | 2.28 Hz | jetpack 7.2.1 · power_mode MAXN_SUPER, clocks pinned with jetson_clocks (GPU 1020 MHz, CPU 1728 MHz) · tensorrt 10.16.2.10conditionsboard: 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) | TensorRT | fp16 | 3 @640x480 | 643.39 / 644.92 | 1.55 Hz | jetpack 7.2.1 · power_mode MAXN_SUPER, clocks pinned with jetson_clocks (GPU 1020 MHz, CPU 1728 MHz) · tensorrt 10.16.2.10conditionsboard: 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.
[](https://caniembody.pages.dev/compat/i2rt-yam/pi05)<iframe src="https://caniembody.pages.dev/embed/i2rt-yam/pi05" width="380" height="132" style="border:0" title="π0.5 on I2RT YAM | CanIEmbody"></iframe>History
- 2026-10-25 fecccd1 Initial release: robot policy compatibility matrix · Danilo Znamerovszkij