mcMotionChallengeGitHub ↗

EVALUATION RECORD

MOSAIC GMT

LAFAN1 × G1 29DoF · Ideal proprioception · no global position

Evaluated by MotionChallenge using the shared protocol.

Results

11 / 40motions passedCompletion 54.64%

Success and completion use the first declared failure. MPBPE measures position errors over the benchmark’s declared body origins, with three alignments over two time ranges.

Mean per-body position error · MPBPE · mm · lower is better
Entire motionUntil first failure
XY onlyXY + yawNo alignmentXY onlyXY + yawNo alignment
305.64289.472587.2063.8262.711277.90

Entire-motion errors include every frame through the endpoint, including post-failure states. The other range ends at the first failing sample, including that sample.

Benchmark
LAFAN1 × G1 29DoF · Ideal proprioception · no global position
Engine
MuJoCo 3.17.0 · official CPU batch
Checkpoint SHA256
d6e124126a1b3e4fd3cc75e820fc0a4e170aa66ab297290fa1a3f1495fa07f3d
Evaluation code
MotionChallenge-mosaic
Core
MotionChallenge-Core
Verification
Both trajectory ranges recomputed · repeatability checked
Inference comparison
160 inputs · 4 resets · Report ↗ · Comparison code ↗

Reproduce this run

git clone --recurse-submodules https://github.com/MotionChallenge-Project/MotionChallenge-mosaic.git
cd MotionChallenge-mosaic
uv sync
uv run evaluate.py

Runs the full benchmark and verifies the saved trajectories.

Complete results, source & verification (JSON) ↗