Does X-VLA run on WidowX 250 S?
Capabilities
| Capability | State | Notes | Evidence |
|---|---|---|---|
| Train / fine-tune | works | ev-xvla-paper-widowx, ev-xvla-readme-widowx | |
| Real-robot inference | works | Real-world WidowX pick-and-place, five tasks (visual/motion/physical/semantic generalization), 10 trials each. | ev-xvla-paper-widowx, ev-xvla-paper-realworld |
| Simulation inference | works | SimplerEnv WidowX (simulation). | ev-xvla-readme-widowx, ev-lerobot-xvla-doc |
| Teleop + record | no verified evidence | — |
Caveats
- The paper says only 'WidowX' following the BridgeData V2 benchmark. It identifies the arm through BridgeData V2 (a WidowX 250), not by model name.
- X-VLA's repo provides an inference server but no WidowX robot client. Its released WidowX checkpoints (2toINF/X-VLA-WidowX, lerobot/xvla-widowx) are labelled for the SimplerEnv benchmark, and LeRobot has no WidowX 250 robot type.
Evidence 4 sources; the cell shows the highest tier
- T1 · upstream documentedpapersource 2025-10-11verified 2026-10-24 (1d ago)
X-VLA paper, Appendix J: X-VLA-0.9B fine-tuned on BridgeData is deployed on a real WidowX for pick-and-place, with each task evaluated 10 times.
“WidowX for pick-and-place experiments. Specifically, X-VLA-0.9B finetuned on BridgeData is directly deployed to evaluate its ability to perform robust manipulation on a compact platform.”
https://arxiv.org/abs/2510.10274 - T1 · upstream documentedpapersource 2025-10-11verified 2026-10-24 (1d ago)
X-VLA paper, Sec. 5: real-world WidowX evaluation follows the BridgeData V2 benchmark (Fig. 7).
“Real-world experiments. We also evaluate X-VLA-0.9B on physical robotic platforms follow the BridgeDatav2 benchmark (Walke et al., 2023)”
https://arxiv.org/abs/2510.10274 - T1 · upstream documentedofficial-docssource 2026-10-20verified 2026-10-24 (1d ago)
X-VLA README lists the 2toINF/X-VLA-WidowX checkpoint (fine-tuned on BridgeData V2, 95.8% on Simpler) and a server-client inference API.
“Fine-tuned on BridgeDataV2 (Simpler benchmark).”
https://github.com/2toinf/X-VLA/blob/6c4ecfb8be458ae7c0c96094e4b0add4174f4fe2/README.md - T1 · upstream documentedofficial-docssource 2026-08-03verified 2026-10-24 (1d ago)lerobot 0.6.1
LeRobot v0.6.1 X-VLA doc lists lerobot/xvla-widowx under 'Simulation Checkpoints'.
“Fine-tuned on BridgeData for pick-and-place experiments on compact WidowX platforms. Demonstrates robust manipulation capabilities.”
https://github.com/huggingface/lerobot/blob/v0.6.1/docs/source/xvla.mdx
X-VLA inference speed by hardware robot-independent · inference cost only
| Compute | Runtime | Precision | Cams | p50 / p95 ms | Inference | Platform | Source |
|---|---|---|---|---|---|---|---|
| Jetson Orin Nano 8GB (Super) | TensorRT | fp16 | 3 @640x480 | 342.59 / 343.47 | 2.92 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 resized by the exported policy to 224x224; 10 denoising steps; latency is runtime-only over a pre-materialized observation ring. Whole-board RAM in use 3.04 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) | ONNX (split) | fp16 | 3 | 391.55 / 407.33 | 2.55 Hz | jetpack 7.2 (L4T R39.2.x, per the repo host-setup doc at this commit) · power_mode MAXN_SUPER, pinned clocks · cuda 13.2 · tensorrt 10.16.2 (system, via ONNX Runtime TensorRT EP)conditionsboard: Jetson Orin Nano Super 8 GB developer kit · os: Ubuntu 24.04 · onnxruntime-gpu: 1.24 Inference cost only; not task success. Deterministic synthetic observations, pinned MAXN_SUPER clocks. Configured EP priority 12 graphs TensorRT. System RAM 5395 MB in use. | T3 · community reported |
| Jetson Orin Nano 8GB (Super) | PyTorch eager | fp32 | 3 | 2313.5 / 2320.89 | 0.43 Hz | jetpack 7.2 (L4T R39.2.x, per the repo host-setup doc at this commit) · power_mode MAXN_SUPER, pinned clocks · cuda 13.2 · lerobot 0.6.1conditionsboard: Jetson Orin Nano Super 8 GB developer kit · os: Ubuntu 24.04 Inference cost only; not task success. Deterministic synthetic observations, pinned MAXN_SUPER clocks. Stock LeRobot X-VLA policy in PyTorch FP32 (LeRobot 0.6.1 env). At 30 Hz control the 30-step chunk is overrun (author's control-rate fit). System RAM 5450 MB in use. | T3 · community reported |
Embed this cell
Badge for READMEs and model cards. It always shows the current state.
[](https://caniembody.pages.dev/compat/widowx-250/xvla)<iframe src="https://caniembody.pages.dev/embed/widowx-250/xvla" width="380" height="132" style="border:0" title="X-VLA on WidowX 250 S | CanIEmbody"></iframe>History
- 2026-10-25 fecccd1 Initial release: robot policy compatibility matrix · Danilo Znamerovszkij