Cyborg Index · July 2026

Beacon-15 vs Cirrus-05

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

Beacon-15

Vendor Nu

2.90of 5

Rank

Coverage

35/40

Proof

2.04

Rank up 7 to 5 (was 12)

Cirrus-05

Vendor Echo

2.71of 5

Rank

Coverage

33/40

Proof

2.03

Rank down 3 to 7 (was 4)

The short answer

Beacon-15 scores 0.19 higher than Cirrus-05, and independent evidence backs it.

70% of Beacon-15's score is carried by independently verified material rather than claims alone. Cirrus-05 has more unscored ground than the gap between them, so this could change next month.

Where Beacon-15 is ahead

Manipulation+0.59Independently backed
Autonomy+0.58Independently backed
Safety+0.29Independently backed
Learning+0.27Independently backed

Where Cirrus-05 is ahead

HRI+0.35Independently backed

What would change it

7 sub-axes on the trailing side have no public evidence yet. If all of them scored full marks they would be worth 0.88 — against a gap of 0.19.

Locomotion · Stairsup to +0.125
Perception · Scene understandingup to +0.125
Safety · E stop archup to +0.125
Perception · Spatial memoryup to +0.125
Deployment · Customer diversityup 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

85%of sub-axes covered

We have public evidence for 85% of the 40 sub-axes across these two.

Beacon-15 · 88%

Cirrus-05 · 83%

The shape of each score

Beacon-152.90

35/40 sub-axes have evidence

Faint outline = Cirrus-05

Cirrus-052.71

33/40 sub-axes have evidence

Faint outline = Beacon-15

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

Manipulation0.59 Autonomy0.58 HRI0.35 Safety0.29 Learning0.27 Perception0.10 Deploymentlevel Locomotionlevel

Beacon-15Cirrus-05

Shared 0–5 track · widest gap first

Manipulation

3.22 / 2.63◄ 0.59

Independently backedEvidenced cells · 3/5 vs 5/5

Grasp variety

3.40 / 2.65◄ 0.75

Bimanual

3.39 / 2.58◄ 0.81

Force control

2.73 / 2.96► 0.23

In hand reorient

3.41 / 2.37◄ 1.04

Tool use

3.15 / 2.57◄ 0.58

Autonomy

3.09 / 2.51◄ 0.58

Independently backedEvidenced cells · 5/5 vs 5/5

Unattended operation

3.22 / 2.53◄ 0.69

Task horizon

3.06 / 2.11◄ 0.95

Generalization

3.22 / 2.79◄ 0.43

Instruction following

2.55 / 2.56

Recovery from failure

3.40 / 2.54◄ 0.86

HRI

2.85 / 3.20► 0.35

Independently backedEvidenced cells · 5/5 vs 5/5

Speech io

2.39 / 3.38► 0.99

Nl instruction

2.64 / 3.41► 0.77

Intent recognition

3.28 / 3.14◄ 0.14

Safe proximity

2.61 / 3.39► 0.78

Social presentation

3.33 / 2.68◄ 0.65

Safety

2.65 / 2.36◄ 0.29

Independently backedEvidenced cells · 3/5 vs 4/5

Functional safety

3.11 / 2.75◄ 0.36

Certifications

2.26 / 2.36► 0.10

E stop arch

2.41 / 2.03◄ 0.38

Incident transparency

3.06 / 2.47◄ 0.59

Force impact limits

2.42 / 2.21◄ 0.21

Learning

2.90 / 2.63◄ 0.27

Independently backedEvidenced cells · 4/5 vs 5/5

Imitation learning

2.68 / 3.07► 0.39

Sim to real

3.34 / 2.27◄ 1.07

Foundation policy

3.13 / 2.13◄ 1.00

Data flywheel

2.52 / 3.06► 0.54

Transfer fine tune

2.81 / 2.62◄ 0.19

Perception

3.01 / 2.90◄ 0.10

Independently backedEvidenced cells · 5/5 vs 2/5

Sensor stack

3.28 / 2.83◄ 0.45

Scene understanding

2.90 / 3.37► 0.47

Spatial memory

2.93 / 3.09► 0.16

Object permanence

3.11 / 2.44◄ 0.67

Semantic world model

2.81 / 2.78

Deployment

2.54 / 2.50

Not enough public record yetEvidenced cells · 5/5 vs 4/5

Named pilots

2.33 / 1.97◄ 0.36

Field hours

2.75 / 2.06◄ 0.69

Customer diversity

2.40 / 2.87► 0.47

Supportability

2.37 / 2.87► 0.50

Commercial posture

2.87 / 2.73◄ 0.14

Locomotion

2.94 / 2.96

Not enough public record yetEvidenced cells · 5/5 vs 3/5

Flat ground walking

2.67 / 3.06► 0.39

Uneven terrain

3.49 / 2.86◄ 0.63

Stairs

3.02 / 3.02

Disturbance recovery

2.71 / 2.96► 0.25

Speed endurance

2.80 / 2.90► 0.10

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.

Beacon-15

Beacon-15 scores higher on Deployment and Manipulation — 2.77 against 2.54 for Cirrus-05.

Running without a human in the loop

Unattended operation and task horizon, backed by what the robot can perceive.

Beacon-15

Beacon-15 scores higher on Autonomy and Perception — 3.06 against 2.64 for Cirrus-05.

Moving through difficult ground

Uneven terrain, stairs and disturbance recovery.

Level

Beacon-15 and Cirrus-05 read the same on Locomotion (2.94 vs 2.96).

Working alongside people

Functional safety and certification, plus how it reads and answers people.

Level

Beacon-15 and Cirrus-05 read the same on Safety and HRI (2.72 vs 2.64).

Picking up new tasks

Imitation learning, transfer and generalisation across tasks.

Beacon-15

Beacon-15 scores higher on Learning and Autonomy — 2.96 against 2.59 for Cirrus-05.

Research and developer use

How much of the stack is documented and reproducible.

Beacon-15

Beacon-15 scores higher on Learning and Perception — 2.93 against 2.72 for Cirrus-05.

How solid the record is on each side

Beacon-15

Sub-axes on the record: 35 of 40 · 88%
Carried by verified material: 70%
Rests on claims and demos: 0.86

Best documented — Locomotion, Manipulation, Perception, Autonomy, Learning, HRI, Deployment, Safety

Nothing on the record — every axis has at least one read

Cirrus-05

Sub-axes on the record: 33 of 40 · 83%
Carried by verified material: 75%
Rests on claims and demos: 0.68

Best documented — Locomotion, Manipulation, Autonomy, Learning, HRI, Deployment, Safety

Nothing on the record — every axis has at least one read

What this comparison cannot settle

12 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