Vision Language Action Python Packages

inspect-robots

Open source evals for physical AI. Run any LLM/VLA on any arm/humanoid against any real/sim benchmark.

robot-reel

Review Physical AI with original evidence: compare VLM explanations against recorded robot outcomes, replay SmolVLA failures, and inspect Newton/Genesis simulations offline.

gauntlet-robotics

Robustness evaluation harness for robot manipulation policies (VLA): perturbation sweeps, failure clustering, reproducible reports.

provael

Provael — red-team open Vision-Language-Action (VLA) robot policies in simulation and report an Attack Success Rate (ASR). Prove it. Prevail.

fieldtrial

Statistically rigorous real-world evaluation for robot policies. Plan how many rollouts you need, run them randomized and blinded from a phone at the robot, let fieldtrial switch LeRobot or openpi policies itself, and get a report that only says "better" when a pre-registered test does.

strands-robots

Control robots and physical hardware with natural language through Strands Agents.

inspect-robots-agent

Open source evals for physical AI. Run any LLM/VLA on any arm/humanoid against any real/sim benchmark.

inspect-robots-capx

Open source evals for physical AI. Run any LLM/VLA on any arm/humanoid against any real/sim benchmark.

mlx-smolvla

SmolVLA on Apple Silicon: native MLX inference, LeRobot-protocol serving, and preview on-device training, with verified parity gates

inspect-robots-voice

Open source evals for physical AI. Run any LLM/VLA on any arm/humanoid against any real/sim benchmark.

inspect-robots-isaacsim

Open source evals for physical AI. Run any LLM/VLA on any arm/humanoid against any real/sim benchmark.

strands-robots-sim

Simulated environments for robot agent evaluation and reinforcement learning.

inspect-robots-xpolicylab

Open source evals for physical AI. Run any LLM/VLA on any arm/humanoid against any real/sim benchmark.

inspect-robots-ros

Open source evals for physical AI. Run any LLM/VLA on any arm/humanoid against any real/sim benchmark.

wamprobe

Counterfactual evaluation for World Action Models

tetherfast

FastCrest Tether: the OSS edge-to-cloud AI deploy CLI. Optimize, verify, deploy across Jetson, RTX, Apple Silicon, AMD. Hybrid edge-cloud inference with parity certs.

tether-cli

FastCrest Tether: the OSS edge-to-cloud AI deploy CLI. Optimize, verify, deploy across Jetson, RTX, Apple Silicon, AMD. Hybrid edge-cloud inference with parity certs.