About humanoidmetric.com
Why it exists · what it refuses to do · what makes it hard to copyhumanoidmetric.com is an independent benchmark that scores robots 0–100 from published, cited specifications — 188 machines from 100 companies across 20 countries, built on 570 source citations. It is not funded by, affiliated with, or reviewed by any robotics manufacturer. Nothing on it is a matter of opinion that isn't shown as one.
Why we built it
The humanoid-robot field is drowning in demo videos and valuation headlines. Anyone trying to answer a simple question — which of these machines actually works, and how would I know? — runs into four failures that no amount of reading press coverage fixes.
1 · A demo doesn't show autonomy
A robot folding laundry on stage may be running its own policy or may be a puppet worn by a person in a motion-capture suit in the next room. The footage looks identical. The distinction is the entire question — it separates a product from a prototype — and it is almost never in the caption.
So autonomy is the heaviest pillar here at 25%, scored from a published ladder rather than an impression, and teleoperation is capped near the floor of it: research 20, teleoperated 35, supervised 70, task-autonomous 92. No profile for any robot category is permitted to redefine those rungs.
2 · Spec sheets are marketing documents
Payloads are quoted at best case, runtimes without load, and the numbers that would embarrass a machine are simply absent. A ranking that fills the gaps with estimates quietly rewards whoever discloses least.
Records retain cited sources, a confidence grade and an as_of date.
Field reviews are shown separately; legacy baselines may be incomplete.
Domain diversity and model agreement alone do not establish independent observation.
Missing specifications are excluded from weighted sub-metrics and evidence coverage caps the pillar.
3 · One score can't span every robot
The obvious move — one grand index over every robot — produces nonsense the moment you leave humanoids. Score a quadruped against a human hand's 22 degrees of freedom and manipulation returns 0.0 for lacking thumbs. Score a surgical arm on walking speed and mobility returns 0.0 for being bolted to the floor. Both are arithmetically correct and semantically worthless.
So each category carries its own index — its own weights, its own reference scales — and composites are ranked within a category, never across. The 10 categories →
4 · Rankings go stale faster than they're rebuilt
This field ships something material every month. A hand-built ranking is substantially wrong within a quarter, and the ones you find are usually a blog post with a date on it and no way to tell what changed since.
So this one is continuously re-researched, every change is checked against its sources before it lands, and every change that lands is kept in the record's public version history: each field that changed, old → new, and when. You can see what moved, when, and from which sources on each robot page.
Kept current, not rebuilt
Most rankings are a snapshot: assembled once, accurate on publication day, quietly wrong six months later. This one is re-researched continuously, and every change to it is published with the sources it was checked against.
New automatic updates pass citation and corroboration checks. Source origin and field reviews remain visible; older ranked records may retain incomplete review. Unsupported proposed values are held for review, and absent specifications are left blank. The full record of what changed and when is in each robot's and company's version history.
The second structural bet: one index per category
Why the taxonomy is load-bearing
Most benchmarks that expand do it by adding rows. Adding a drone to a humanoid leaderboard produces a number that is arithmetically valid and means nothing, so this platform expands by adding indices instead — 10 live scoring profiles across 10 declared categories, each with its own weights and its own reference scales, anchored to that category's frontier rather than a human body.
A profile that omits a pillar is saying not applicable to this category, not "scores zero" — the remaining weights are renormalised, so a quadruped is not docked a fifth of its index for lacking thumbs. And a category with no registered profile is refused at every admission gate: a record filed there would silently fall back to being scored as a humanoid, which is acceptable for rendering an old row and unacceptable for admitting a new one.
What survives across categories
Exactly three pillars, and the reason is worth stating precisely: they are scored from ladders rather than physical measurements. A teleoperated submarine and a teleoperated humanoid are equally teleoperated; shipping means shipping; money is money.
- Autonomy — the same four rungs everywhere, teleop capped in all of them
- Commercial readiness — production maturity plus named, cited deployments
- Ecosystem & viability — funding, valuation, corporate backing
That makes a cross-category autonomy comparison honest where a cross-category composite never could be. The published indices today:
How to check us
Every claim on this site is traceable
- The methodology renders each scoring profile verbatim from the code that computes it — weights, ladders, reference scales and what each denominator is anchored to.
- Each robot page lists its source URLs, its confidence grade, and the date its figures were last verified.
- The full dataset is available as JSON with a free API key. Recompute the index yourself and tell us where we're wrong.
- Every robot and company page keeps a version history: each field that changed, old → new, and whether a research update or a direct correction made it.
- All 169 published humanoid scores are locked by a snapshot test, so a change to the scoring code that silently moves a robot fails the build instead of shipping.
What this is not
- Not a buyer's guide. A high index score means a machine is well documented and capable on published evidence — not that it suits your process, your safety case or your budget.
- Not investment advice. The ecosystem pillar reads funding and valuation as signals of survival odds, nothing more.
- Not vendor-funded, and not affiliated with any manufacturer. No company pays to appear, to rank, or to be re-checked, and no listing is removable by request — corrections are, with sources. See the trademark notice.
- Not a measure of what a robot could do. It scores what has been demonstrated and published. Unreleased capability is real and is not evidence.
Corrections
If a figure here is wrong, we want the correction more than we want to have been right — including from the manufacturer. Send the claim and at least one primary source to governance@autogovern.io. It enters the same review queue as everything else, gets the same validity checks, and if it's applied it appears in that record's version history like any other change. Operated by autogovern.io · legal.