Cyborg Index · July 2026
Kepler Forerunner K2 vs NEURA 4NE-1
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-axesThe short answer
Kepler Forerunner K2 scores 0.32 higher than NEURA 4NE-1, but that lead rests mostly on vendor claims and demos.
Only 5% of Kepler Forerunner K2's score is carried by independently verified material — the rest is claimed or shown in demos. NEURA 4NE-1 has more unscored ground than the gap between them, so this could change next month.
Where Kepler Forerunner K2 is ahead
- Locomotion+1.20Rests on claims and demos
- Deployment+0.80Rests on claims and demos
- Manipulation+0.60Rests on claims and demos
- Perception+0.40Rests on claims and demos
Where NEURA 4NE-1 is ahead
- Safety+0.40Rests on claims and demos
What would change it
11 sub-axes on the trailing side have no public evidence yet. If all of them scored full marks they would be worth 1.38 — against a gap of 0.32.
- Learning · Foundation policyup to +0.125
- Perception · Semantic world modelup to +0.125
- Learning · Imitation learningup to +0.125
- Manipulation · Bimanualup to +0.125
- Locomotion · Uneven terrainup to +0.125
There is more unscored ground than there is gap, so this result could turn over after the next research month.
How much we know
60%of sub-axes covered
We have public evidence for 60% of the 40 sub-axes across these two.
Kepler Forerunner K2 · 65%
NEURA 4NE-1 · 55%
The shape of each score
26/40 sub-axes have evidence
Faint outline = NEURA 4NE-1
22/40 sub-axes have evidence
Faint outline = Kepler Forerunner K2
Reading the wheels — clockwise from the top
- 1 Locomotion
- 2 Manipulation
- 3 Perception
- 4 Autonomy
- 5 Learning
- 6 HRI
- 7 Deployment
- 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
Kepler Forerunner K2NEURA 4NE-1
Shared 0–5 track · widest gap firstLocomotion
2.00 / 0.80◄ 1.20
Rests on claims and demosEvidenced cells · 4/5 vs 2/5
Flat ground walking
2.00 / 2.00
Uneven terrain
3.00 / —◄ 3.00
Stairs
— / —
Disturbance recovery
3.00 / —◄ 3.00
Speed endurance
2.00 / 2.00
Deployment
1.20 / 0.40◄ 0.80
Rests on claims and demosEvidenced cells · 4/5 vs 1/5
Named pilots
1.00 / —◄ 1.00
Field hours
1.00 / —◄ 1.00
Customer diversity
3.00 / —◄ 3.00
Supportability
— / —
Commercial posture
1.00 / 2.00► 1.00
Manipulation
1.80 / 1.20◄ 0.60
Rests on claims and demosEvidenced cells · 3/5 vs 3/5
Grasp variety
3.00 / 2.00◄ 1.00
Bimanual
4.00 / —◄ 4.00
Force control
2.00 / 2.00
In hand reorient
— / —
Tool use
— / 2.00► 2.00
Safety
0.20 / 0.60► 0.40
Rests on claims and demosEvidenced cells · 1/5 vs 3/5
Functional safety
1.00 / 2.00► 1.00
Certifications
— / 0.00
E stop arch
— / —
Incident transparency
— / —
Force impact limits
— / 1.00► 1.00
Perception
1.40 / 1.00◄ 0.40
Rests on claims and demosEvidenced cells · 4/5 vs 3/5
Sensor stack
2.00 / 2.00
Scene understanding
1.00 / 1.00
Spatial memory
2.00 / 2.00
Object permanence
— / —
Semantic world model
2.00 / —◄ 2.00
Autonomy
1.20 / 1.20
Not enough public record yetEvidenced cells · 4/5 vs 3/5
Unattended operation
0.00 / 2.00► 2.00
Task horizon
3.00 / 2.00◄ 1.00
Generalization
1.00 / —◄ 1.00
Instruction following
2.00 / 2.00
Recovery from failure
— / —
Learning
1.00 / 1.00
Not enough public record yetEvidenced cells · 2/5 vs 3/5
Imitation learning
2.00 / —◄ 2.00
Sim to real
— / 1.00► 1.00
Foundation policy
3.00 / —◄ 3.00
Data flywheel
— / 2.00► 2.00
Transfer fine tune
— / 2.00► 2.00
HRI
1.80 / 1.80
Not enough public record yetEvidenced cells · 4/5 vs 4/5
Speech io
2.00 / 2.00
Nl instruction
3.00 / 2.00◄ 1.00
Intent recognition
— / 2.00► 2.00
Safe proximity
2.00 / 3.00► 1.00
Social presentation
2.00 / —◄ 2.00
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.
Kepler Forerunner K2
Kepler Forerunner K2 scores higher on Deployment and Manipulation — 1.40 against 0.67 for NEURA 4NE-1.
Running without a human in the loop
Unattended operation and task horizon, backed by what the robot can perceive.
Kepler Forerunner K2
Kepler Forerunner K2 scores higher on Autonomy and Perception — 1.27 against 1.13 for NEURA 4NE-1.
Moving through difficult ground
Uneven terrain, stairs and disturbance recovery.
Kepler Forerunner K2
Kepler Forerunner K2 scores higher on Locomotion — 2.00 against 0.80 for NEURA 4NE-1.
Working alongside people
Functional safety and certification, plus how it reads and answers people.
NEURA 4NE-1
NEURA 4NE-1 scores higher on Safety and HRI — 1.00 against 0.73 for Kepler Forerunner K2.
Picking up new tasks
Imitation learning, transfer and generalisation across tasks.
Level
Kepler Forerunner K2 and NEURA 4NE-1 read the same on Learning and Autonomy (1.07 vs 1.07).
Research and developer use
How much of the stack is documented and reproducible.
Kepler Forerunner K2
Kepler Forerunner K2 scores higher on Learning and Perception — 1.13 against 1.00 for NEURA 4NE-1.
How solid the record is on each side
Kepler Forerunner K2
- Sub-axes on the record
- 26 of 40 · 65%
- Carried by verified material
- 5%
- Rests on claims and demos
- 1.26
Best documented — Locomotion, Manipulation, Perception, Autonomy, HRI, Deployment
Nothing on the record — every axis has at least one read
NEURA 4NE-1
- Sub-axes on the record
- 22 of 40 · 55%
- Carried by verified material
- 8%
- Rests on claims and demos
- 0.92
Best documented — Manipulation, Perception, Autonomy, Learning, HRI, Safety
Nothing on the record — every axis has at least one read
What this comparison cannot settle
- Neither humanoid has a public read on 7 of the 40 sub-axes, so those cannot be compared at all.
- 18 sub-axes are read on only one side. A one-sided cell is undecided, not a zero for the silent side.
- Most of Kepler Forerunner K2's score rests on vendor statements and demo footage rather than independently verified material (5% carried by verified evidence).
- Most of NEURA 4NE-1's score rests on vendor statements and demo footage rather than independently verified material (8% carried by verified evidence).
- 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.