CYBORGS OF EARTH
Cyborg Index · July 2026

Halide-03 vs Nimbus-01

Both humanoids read on the same 8 axes and 40 sub-axes, using the published Index scores. Every difference below carries the public evidence behind it, and sub-axes with no public evidence count as 0 rather than being skipped.

Overall standing

Capability score · 0–5 over 40 sub-axes
Halide-03
Vendor Charlie
3.35of 5
Rank
#2
Coverage
40/40
Proof
1.95
Rank up 3 to 2 (was 5)
Nimbus-01
Vendor Alpha
3.48of 5
Rank
#1
Coverage
36/40
Proof
2.72
Rank up 1 to 1 (was 2)
The short answer

Nimbus-01 scores 0.12 higher than Halide-03, and independent evidence backs it.

78% of Nimbus-01's score is carried by independently verified material rather than claims alone. Halide-03 cannot close the gap on unscored ground alone.

Where Halide-03 is ahead
  • Learning+1.19Rests on claims and demos
  • Autonomy+0.96Rests on claims and demos
  • Perception+0.56Independently backed
  • Manipulation+0.36Independently backed
Where Nimbus-01 is ahead
  • Deployment+2.26Independently backed
  • HRI+1.06Independently backed
  • Safety+0.46Independently backed
  • Locomotion+0.29Independently backed
What would change it

There are no blank sub-axes on the trailing side, so this gap can only move if new evidence changes a score that already exists.

How much we know
95%of sub-axes covered

We have public evidence for 95% of the 40 sub-axes across these two.

Halide-03 · 100%
Nimbus-01 · 90%
The shape of each score
Halide-033.35
LocomotionManipulationPerceptionAutonomyLearningHRIDeploymentSafety

40/40 sub-axes have evidence

Faint outline = Nimbus-01

Nimbus-013.48
LocomotionManipulationPerceptionAutonomyLearningHRIDeploymentSafety

36/40 sub-axes have evidence

Faint outline = Halide-03

Reading the wheels — clockwise from the top
  1. 1 Locomotion
  2. 2 Manipulation
  3. 3 Perception
  4. 4 Autonomy
  5. 5 Learning
  6. 6 HRI
  7. 7 Deployment
  8. 8 Safety

Each wheel has 40 spokes — five per axis. A longer spoke is a higher score, and the dashed rings mark 1 to 4. Hover or tab onto any spoke to read the axis, the sub-axis and the exact value. A missing spoke means there is no public evidence for that sub-axis yet, which is not the same as scoring zero. The faint outline on each wheel is the other humanoid, drawn on the same scale.

Every axis, every sub-axis
Deployment2.26Learning1.19HRI1.06Autonomy0.96Perception0.56Safety0.46Manipulation0.36Locomotion0.29
Halide-03Nimbus-01
Shared 0–5 track · widest gap first
Deployment
2.06 / 4.32► 2.26
Independently backedEvidenced cells · 5/5 vs 5/5
Named pilots
2.48 / 4.46► 1.98
Field hours
2.30 / 4.07► 1.77
Customer diversity
2.25 / 3.91► 1.66
Supportability
1.70 / 4.54► 2.84
Commercial posture
1.56 / 4.61► 3.05
Learning
4.09 / 2.89◄ 1.19
Rests on claims and demosEvidenced cells · 5/5 vs 5/5
Imitation learning
4.19 / 2.71◄ 1.48
Sim to real
4.14 / 3.20◄ 0.94
Foundation policy
3.79 / 2.77◄ 1.02
Data flywheel
4.17 / 2.99◄ 1.18
Transfer fine tune
4.15 / 2.80◄ 1.35
HRI
2.61 / 3.68► 1.06
Independently backedEvidenced cells · 5/5 vs 5/5
Speech io
2.46 / 3.71► 1.25
Nl instruction
2.44 / 3.21► 0.77
Intent recognition
2.53 / 3.93► 1.40
Safe proximity
2.97 / 4.03► 1.06
Social presentation
2.66 / 3.50► 0.84
Autonomy
3.50 / 2.54◄ 0.96
Rests on claims and demosEvidenced cells · 5/5 vs 5/5
Unattended operation
3.17 / 2.38◄ 0.79
Task horizon
3.76 / 2.81◄ 0.95
Generalization
3.61 / 2.38◄ 1.23
Instruction following
3.93 / 2.34◄ 1.59
Recovery from failure
3.03 / 2.77◄ 0.26
Perception
3.87 / 3.31◄ 0.56
Independently backedEvidenced cells · 5/5 vs 4/5
Sensor stack
4.14 / 3.12◄ 1.02
Scene understanding
3.56 / 3.41◄ 0.15
Spatial memory
3.99 / 3.47◄ 0.52
Object permanence
4.01 / 3.45◄ 0.56
Semantic world model
3.66 / 3.10◄ 0.56
Safety
2.99 / 3.45► 0.46
Independently backedEvidenced cells · 5/5 vs 3/5
Functional safety
2.68 / 3.61► 0.93
Certifications
2.88 / 3.35► 0.47
E stop arch
3.06 / 3.38► 0.32
Incident transparency
3.16 / 3.46► 0.30
Force impact limits
3.19 / 3.46► 0.27
Manipulation
4.08 / 3.72◄ 0.36
Independently backedEvidenced cells · 5/5 vs 5/5
Grasp variety
4.07 / 3.89◄ 0.18
Bimanual
3.97 / 4.15► 0.18
Force control
3.81 / 3.63◄ 0.18
In hand reorient
4.46 / 3.67◄ 0.79
Tool use
4.07 / 3.26◄ 0.81
Locomotion
3.63 / 3.93► 0.29
Independently backedEvidenced cells · 5/5 vs 4/5
Flat ground walking
3.11 / 4.05► 0.94
Uneven terrain
3.69 / 3.71
Stairs
3.51 / 4.04► 0.53
Disturbance recovery
4.08 / 3.93◄ 0.15
Speed endurance
3.78 / 3.90► 0.12
Segment length and thickness are the gapNo public evidence yet
Which is better for what

Each row is decided by the axes that matter for that kind of work, read off the same Capability scores as the ranking. Rows only appear where both humanoids have enough public record to answer.

Manufacturing and warehouse work
Named pilots and field hours, backed by what the hands can actually do.
Nimbus-01
Nimbus-01 scores higher on Deployment and Manipulation — 4.12 against 2.73 for Halide-03.
Running without a human in the loop
Unattended operation and task horizon, backed by what the robot can perceive.
Halide-03
Halide-03 scores higher on Autonomy and Perception — 3.62 against 2.79 for Nimbus-01.
Moving through difficult ground
Uneven terrain, stairs and disturbance recovery.
Nimbus-01
Nimbus-01 scores higher on Locomotion — 3.93 against 3.63 for Halide-03.
Working alongside people
Functional safety and certification, plus how it reads and answers people.
Nimbus-01
Nimbus-01 scores higher on Safety and HRI — 3.53 against 2.87 for Halide-03.
Picking up new tasks
Imitation learning, transfer and generalisation across tasks.
Halide-03
Halide-03 scores higher on Learning and Autonomy — 3.89 against 2.77 for Nimbus-01.
Research and developer use
How much of the stack is documented and reproducible.
Halide-03
Halide-03 scores higher on Learning and Perception — 4.02 against 3.03 for Nimbus-01.
How solid the record is on each side
Halide-03
Sub-axes on the record
40 of 40 · 100%
Carried by verified material
58%
Rests on claims and demos
1.40
Best documented — Locomotion, Manipulation, Perception, Autonomy, Learning, HRI, Deployment, Safety
Nothing on the record — every axis has at least one read
Nimbus-01
Sub-axes on the record
36 of 40 · 90%
Carried by verified material
78%
Rests on claims and demos
0.76
Best documented — Locomotion, Manipulation, Perception, Autonomy, Learning, HRI, Deployment, Safety
Nothing on the record — every axis has at least one read

What this comparison cannot settle

  • 4 sub-axes are read on only one side. A one-sided cell is undecided, not a zero for the silent side.
  • Demonstrations happen in different environments and product revisions are not always equivalent, so like-for-like uptime and throughput figures do not exist for most of the field.

Go deeper on either humanoid

Nearby comparisons

All comparisons →
Cyborg IndexJuly 2026 editionCapability scoremethodologypreview cohort