Physical plant
29 source joints
All leg, waist, and arm bodies carry the pinned mode-11 inertias, joint axes, and hard limits.
Teach a whole-body G1 model to walk
Twenty-nine source actuators, 480 Hz articulated dynamics, and 5,040 learned locomotion weights — optimized live in your browser by CMA-ES with no gradient ever computed.
Learned controller
15 × 42 × 8
Fifteen locomotion rows multiply 42 physical signals by eight periodic basis terms: 5,040 weights.
Disclosed reflex
14 arm joints
The arms add real mass and reaction forces while a deterministic swing-and-balance reflex drives them.
Interactive Story Mode · Guided Walkthrough
Click a chapter to explore the learning progressionowner poses · 480 Hz terrain physicswalking curriculum mean
Cyan / violet rings are contact booleans. 1928 Sears Craftsman Living Room · Drag to orbit · pinch to zoom.Rose arrow: disclosed lateral push · arm joints are kernel-posed with real mass (head/hands: display-only)
0.0050
0.0002 (refine)0.005 (explore)0.01 (aggressive)
Balanced exploration (default): roughly half the population diverges meaningfully from the mean — the recommended starting point.
1928 Sears Craftsman Bungalow
Honor Bilt Kit ArchitectureParametric architectural reconstruction with 70+ authentic period furnishings & multi-room navigation
Total Corridor Distance: 18.2 m
Full estate traversal from front veranda through parlor, dining, kitchen, hallway, and bedroom suite.
Living Room (Parlor) Physical Environment Profile
Room Area: 23.5 m² · Ceiling Height: 2.85 mCentral gathering room with quartersawn white oak flooring, 1.35m board-and-batten wainscoting, inglenook brick fireplace, and exposed coffered ceiling box beams.
✨ Clinker brick fireplace✨ Coffered ceiling box beams✨ Stickley Morris armchair✨ Dirk van Erp mica lamp
Frankensim G1 flagship
Optimize a 5,040-D walking policy
Fifteen actuators each read 42 physical signals through eight gait-phase basis terms:15 × 42 × 8 = 5,040 learned weights
A disclosed full-CMA curriculum learned 105 meaningful owner coordinates: standing bias, periodic foot unloading, then pelvis feedback. Live search expands that curriculum to all 5,040 weights. Every candidate is scored on the same 1.5-second, 720-step terrain-and-push experiment you watch.
15-Dstand
+90-Dtransfer weight
5,040-Drefine live
Full CMA is implemented on the 128-D arm below, but its O(n²) covariance would contain 25,401,600 entries here; the browser boundary honestly refuses it above 256-D.
The wave field and half-sine lateral shove are deterministic owner inputs. Survival is lexicographically primary: one extra integrated physics step beats every possible shaping-score difference. “Recovery” is horizon-censored when the robot never returns to the disclosed upright band.
Challenge
16 × 16 = 256 candidates
87.5k15k22.5k30k
16 gens is a refinement pass; the curriculum mean itself was learned over hundreds. Every press CONTINUES the same CMA run — mean, sigma, and covariance path preserved — so presses stack: 16 + 2000 + 2000 … up to 30k generations of real search. Wall time scales with your hardware; the HUD shows the live generation. A flat objective stays flat — the run is honest about that too.
Loading the owner-composed G1 experiment…
Multi-Factor Objective Equalizer & Personality Sculptor
How the 11 reward channels sculpt robot behaviorScalable variants, one physical budget
A live 5,040-D terrain-and-push race
Same curriculum mean, Philox seed, population of 16, physical evaluator, and evaluation budget. Full CMA is absent only because the owner correctly refuses dense covariance above 256 dimensions; all four families race on the 128-D arm.
Full CMA-ES
Every covariance interaction
O(n²) storage · O(n³) update
Separable CMA-ES
One variance per coordinate
O(n) storage · O(n) update
LM-CMA
A bounded history of directions
O(mn) storage · O(mn) update
LM-MA
A bounded moving transform
O(mn) storage · O(mn) update
What the kernel actually does (and doesn't)
tap to expand
What the kernel actually does (and doesn't)
Modeled
- · 15 actuated DoFs (legs + waist)
- · Free-floating base, SE(3) poses
- · Semi-implicit Euler, fixed dt = 1/480 s
- · Penalty-based normal contact + Coulomb friction (μ ≈ 0.6)
- · Static Hertz preload at simulation start
- · Five terminal-guard detectors (horizon, height, tilt, contact, joint-limit)
Simplified
- · Four compliant foot patches, not full soles
- · No torso arms, hands, or upper shell
- · No motor torque curves or thermal limits
- · No joint belt-elasticity or backlash
- · Terrain is a 1-D heightfield, not a mesh
- · Policy is a periodic basis, not a neural net
Not modeled
- · No rolling or sliding friction asymmetry
- · No slip detection or recovery reflex
- · No inertial measurement, encoder, or actuator lag
- · No environment wind, vibration, or camera noise
- · No sim-to-real transfer or hardware validation
- · No learned controller beyond the periodic basis
A walker that survives the kernel can still fall on real hardware. The page deliberately stops at a deterministic explainer experiment; treating it as a Unitree validation would be a category error.
Phase prior vs legacy transformer — action-causal transfer check
The flagship above uses a 5,040-D linear residual policy on a hand-designed phase basis — a strong, sample-efficient prior. The committed transformer is a real PPO+Muon artifact, but its policy head is all zero and its superseded training stand-in moved without action. The component below transfers those exact weights without retraining onto the current action-causal stand-in beside a live CMA-ES search. The resulting zero-distance transformer run is an honest measured failure, not a learned-walking claim. Artifacts remain under public/robots/g1/transformer/.
Measuring both policies in a background worker (live CMA-ES search + 720-step transformer rollout) — the page stays interactive.
What this simulation actually does
Every candidate policy is a vector of 5,040 learned weights: 15 lower-body and waist actuators each read 42 physical signals through 8 gait-phase basis terms (15 × 42 × 8 = 5,040). The policy outputs bounded residual efforts; an articulated multibody kernel with SE(3) integration, contact, and friction integrates all 29 source joints at a fixed timestep — the same 1.5-second, 720-step experiment for every candidate and for the winner you watch.
CMA-ES never sees derivatives. It samples a population from a Gaussian search distribution, scores each walk (upright distance, foot contact schedule adherence, energy, and hard guards for falls and joint limits), then reshapes its covariance toward the successful candidates. Full CMA-ES is refused above 256 dimensions because a dense 5,040² covariance would need 25,401,600 entries; the live flagship therefore uses the separable and limited-memory variants you can compare directly.
The source boundary is precise. Frankensim transcribes Unitree's current 29-DoF mode-11 description; the Three.js scene projects the 30 emitted world-from-link poses and never recomputes robot kinematics. The fixed head and hand shells are visual geometry. This remains a deterministic explainer, not a validated hardware controller or sim-to-real result. See the official model guide and its mode-11 URDF.