OpenRAL/rskill-rtdetr-v2-r50vd
Object Detection โข Updated
OpenRAL
non-profit
AI & ML interests
AI, Robotics
Recent Activity
View all activity
Organization Card
OpenRAL
The runtime layer between VLA models and real robots. Open-source, typed, layered โ composable like LangGraph, distributable like Hugging Face.
โฌ GitHub
๐ Docs
๐ฌ Discord
Apache-2.0
ROS 2 Jazzy
Python 3.12
16 robots ยท 20 rSkills ยท 22 benchmarks
A VLA model alone can't run on your robot. It needs a camera pipeline, an observation normaliser, an action de-normaliser, a safety wrapper, a replanning layer when it fails, and a way to log everything for later fine-tuning. OpenRAL is that infrastructure.
- Typed runtime โ eight well-defined layers connected by Pydantic v2 contracts. No magic globals, no hidden retries.
- rSkill format โ Hub repos containing weights, a
rskill.yamlmanifest, quantisation hints, latency budgets, and reproducibleeval/. Install like a model. - Dual-system planning โ a fast visuomotor policy (S1, 30โ200 Hz) and a slow LLM planner (S2) emitting typed tool-calls. Replanning is bounded and explicit.
- Safety kernel โ deny-by-default, Python proposes, C++ disposes.
ROSSafetyViolationis never silently caught.
Get started
# Install (no clone needed)
curl -fsSL https://raw.githubusercontent.com/AdrianLlopart/openral/master/scripts/install.sh | bash
openral doctor
# Browse and install a skill
openral rskill list
openral rskill install OpenRAL/rskill-smolvla-libero
# Run a simulated rollout
just sim-libero
# Deploy to hardware
openral deploy run --config deployments/so100_pickplace.yamlArchitecture
0
HAL
Per-robot adapters โ uniform connect / read_state / write_command Protocol
shipped
1
Sensors
SensorSpec โ ROS 2 streams (RGB-D, F/T, IMU, lidar, tactile)
shipped
2
World State
tf2-aware 30 Hz snapshot; carries lifted detected_objects
shipped
3
rSkill (S1)
Fast visuomotor policy โ VLA, 30โ200 Hz, async action chunks
shipped
4
Reasoning (S2)
Slow LLM planner emitting typed ReasonerToolCall tool-calls
partial
5
WAM
World Action Model โ mental simulation, failure anticipation
planned
6
Safety
C++ separate process, deny-by-default, certifiable, E-stop on fault
partial
7
Observability
OpenTelemetry spans + LeRobotDataset v3 flywheel
shipped
What's in this organization
| Type | What | Count |
|---|---|---|
rskill-* models | Policy rSkills โ VLA weights + manifest + eval | 18 |
rskill-* models | Perception rSkills โ RT-DETR object-detection | 2 |
dataset-* | LeRobotDataset v3 demonstration datasets | growing |
Policy rSkills
Each rSkill is a self-contained Hub repo: rskill.yaml manifest, model.safetensors weights, eval/ results, and a model card with runnable examples. Install with openral rskill install OpenRAL/rskill-<name>.
| rSkill | Backbone | Target robot | License |
|---|---|---|---|
rskill-smolvla-libero |
SmolVLA | Franka Panda | Apache-2.0 |
rskill-smolvla-metaworld |
SmolVLA | Rethink Sawyer | Apache-2.0 |
rskill-smolvla-maniskill-franka |
SmolVLA ร ManiSkill3 | Franka Panda | Apache-2.0 |
rskill-xvla-libero |
xVLA (Florence-2) | Franka Panda | Apache-2.0 |
rskill-act-libero |
ACT | Franka Panda | Apache-2.0 |
rskill-act-aloha |
ACT | ALOHA bimanual | MIT |
rskill-act-aloha-insertion |
ACT (peg insertion) | ALOHA bimanual | MIT |
rskill-diffusion-pusht |
Diffusion Policy | PushT 2-D | Apache-2.0 |
rskill-molmoact2-libero-nf4 |
MolmoAct2 NF4 (~5.5 B) | Franka Panda | Apache-2.0 |
rskill-molmoact2-so101-nf4 |
MolmoAct2 NF4 | SO-101 | Apache-2.0 |
rskill-pi05-libero-nf4 |
ฯ0.5 NF4 | Franka Panda | Research (weights non-Apache) |
rskill-pi05-so101-pickplace-nf4 |
ฯ0.5 NF4 | SO-101 | Research (weights) |
rskill-pi05-openarm-vision-nf4 |
ฯ0.5 NF4 bimanual | OpenArm v2 | Apache-2.0 |
rskill-pi05-robocasa365-human300-nf4 |
ฯ0.5 NF4 | Panda Mobile | Research (weights) |
rskill-rldx1-pt-nf4 |
RLDX-1 foundation (Qwen3-VL-8B, ~6.9 B) | Franka Panda | RLWRLD non-commercial |
rskill-rldx1-ft-libero-nf4 |
RLDX-1 fine-tuned | Franka Panda | RLWRLD non-commercial |
rskill-rldx1-ft-gr1-nf4 |
RLDX-1 (GR1 bimanual) | Fourier GR1 | RLWRLD non-commercial |
rskill-rldx1-ft-simpler-widowx-nf4 |
RLDX-1 (SimplerEnv) | WidowX | RLWRLD non-commercial |
Perception rSkills
Emit ObjectsMetadata (2D detections lifted to 3D via depth) rather than an Action. Plug directly into the openral deploy graph.
| rSkill | Backbone | Notes | License |
|---|---|---|---|
rskill-rtdetr-coco-r18 |
RT-DETR R18 | Lightweight ONNX | Apache-2.0 |
rskill-rtdetr-v2-r50vd |
RT-DETR v2 R50vd | Higher accuracy | Apache-2.0 |
Free for any use
Research-permissive weights (check upstream terms)
Non-commercial โ requires OPENRAL_ACCEPT_NONCOMMERCIAL=1
rSkill manifest format
# rskills/smolvla-libero/rskill.yaml (excerpt)
name: "OpenRAL/rskill-smolvla-libero"
version: "0.1.0"
license: "apache-2.0"
role: "s1"
embodiment_tags: ["franka_panda"]
sensors_required:
- modality: "rgb"
vla_feature_key: "observation.images.camera1"
latency_budget_ms: 120
quantization: nullSupported robots
Hardware + sim
SO-100
SO-101
ALOHA bimanual
Sim โ HW bring-up in progress
Franka Panda
UR5e
UR10e
Sim (eval only)
Unitree H1
Unitree G1
Flexiv Rizon 4
Enactic OpenArm v2
Rethink Sawyer
Fourier GR1
Panda Mobile
Google Robot
WidowX
PushT 2D
Benchmarks
22 benchmark configs across LIBERO, MetaWorld, ManiSkill3, SimplerEnv, RoboCasa, gym-aloha, and gym-pusht. Every rSkill ships reproducible eval/<benchmark>.json you can regenerate locally:
openral benchmark run \
--suite libero_spatial \
--vla smolvla:rskill://OpenRAL/rskill-smolvla-libero
openral benchmark reportObservability & data flywheel
Every skill execution is an OpenTelemetry span โ weights revision pinned, camera frames captured, LLM prompts logged. Traces replay as LeRobotDataset v3 rows, closing the loop from deployment back to training data.
openral dashboard # OTLP receiver at :4318, live trace viewerLicense
Everything in this organization is Apache-2.0 open core. A planned commercial tier (PolyForm SBL 1.0.0) will cover agentic orchestration features โ free for teams under 100 people or <$1M revenue, and free for academic research at any size.
rSkill weights are governed by their upstream licenses (Apache-2.0 / MIT / research-permissive / RLWRLD non-commercial). The loader surfaces the posture at install time.
models 25
OpenRAL/rskill-rtdetr-coco-r18
Object Detection โข Updated
OpenRAL/rskill-xvla-libero
Updated
OpenRAL/rskill-smolvla-metaworld
Updated
OpenRAL/rskill-smolvla-maniskill-franka
Updated
OpenRAL/rskill-smolvla-libero
Updated
OpenRAL/rskill-rldx1-ft-libero-nf4
Updated
OpenRAL/rskill-pi05-so100
Updated
OpenRAL/rskill-diffusion-pusht
Updated
OpenRAL/rskill-act-aloha
Updated
datasets 0
None public yet