Figure pricing

About

Figure is a Sunnyvale, California humanoid-robotics company founded in 2022 by Brett Adcock (previously founder of Archer Aviation and Vettery). Its product is a general-purpose humanoid robot — the current generation is Figure 03 — powered by Helix, an in-house Vision-Language-Action (VLA) model that lets the robot follow natural-language commands and manipulate unfamiliar objects. The pitch is broad: a single humanoid form factor that can be dropped into factories, warehouses, commercial settings, and eventually homes, learning tasks rather than being hard-coded for one.

Two strategic moves define the company. First, in early 2025 Figure ended its collaboration agreement with OpenAI and pivoted to running robot AI entirely in-house — CEO Brett Adcock argued that deploying humanoids at scale required Figure to own the brain, and the company unveiled Helix weeks later. Second, Figure is vertically integrating manufacturing through BotQ, an in-house facility announced March 2025 with a stated first-generation capacity of up to 12,000 humanoids per year, designed so that robots help build robots. Both moves are about owning the full stack — model, hardware, and production — rather than buying any layer.

The money has followed the ambition. Figure raised a $675M Series B at a $2.6B valuation in February 2024 (alongside the since-ended OpenAI deal), then a Series C exceeding $1B at a $39B post-money valuation in September 2025 — roughly a 15x valuation step-up in about 18 months — led by Parkway Venture Capital with NVIDIA, Brookfield, Intel Capital, Microsoft and Jeff Bezos among backers.

Crucially for a pricing blueprint: Figure publishes no pricing of any kind. The website (figure.ai) exposes Figure 03, Helix, Company, News, Careers, Culture and a “Master Plan,” with only a newsletter sign-up and a contact form as calls-to-action — no Buy button, no rate card, no per-robot price; figure.ai/pricing returns a 404. Revenue today comes from milestone-based enterprise pilots, the headline being BMW Manufacturing. So this page documents what is honestly known — the company, its funding, its products, and its sales-led commercial posture — rather than inventing numbers Figure has never published.

Pricing summary : a sales-only humanoid lab selling pilots, not price cards

Figure runs a sales-led, no-public-price commercial model. There is no subscription, no published per-robot or per-robot-hour rate, and no self-serve tier to evaluate. The dimensions, such as they can be observed, are:

• Enterprise deployments — quoted bespoke. Figure’s commercial agreements (e.g. BMW Manufacturing) are described as milestone-based, meaning value and payment are tied to deployment milestones rather than a published meter. No floor price, packaging, or SKU is disclosed; the only contact path is a general inquiry form.
• No robots-as-a-service rate card — some humanoid peers like Agility Robotics publish (or have cited) a per-robot-hour RaaS figure. Figure discloses no such public rate, so even the “rent a robot by the hour” mental model has no Figure number behind it.
• No consumer price — Figure 03 is positioned for the home eventually, and a ~$20,000 target circulates widely, but that figure traces only to third-party analysts. Figure has published nothing official, so it is not a price you can act on.

What makes this different: unlike peers that anchor on a public meter — Waymo’s per-ride fare or a RaaS per-robot-hour rate — Figure keeps its entire commercial surface behind a sales conversation. The only public numbers are funding and valuation, which describe how much investors paid for equity, not how much a customer pays for a robot. For how to think about metering a physical, outcome-shaped deliverable, see choosing the right usage metric.

Pricing by product

Sales motions across products: fully sales-led for the only revenue-bearing surface (enterprise humanoid pilots, quoted via the contact form). There is no self-serve, no free tier, and no public RaaS or per-unit rate — every engagement is scoped and quoted.

Hidden costs : What Figure customers actually pay

Because Figure publishes no price, the “real bill” question can’t be answered from the public site at all — and the costs that would matter are exactly the ones a sales conversation reveals.

Path 1 — an enterprise deployment. Everything is in the quote. With no published floor, no packaging tiers, and no worked examples, a prospective operator cannot estimate cost without engaging sales. The likely cost drivers, by analogy to other physical-automation deployments, are the robots themselves (purchase or RaaS), integration and safety engineering to fit the robot into a real line, Helix/software and updates, and ongoing support/maintenance — none of which carry a public number. The hidden cost is evaluation friction: there is no calculator a buyer can self-serve against.

Path 2 — waiting for a home robot. Figure 03 is positioned for the home, but it is not yet commercially available (alpha home testing was targeted for late 2025; broader availability not before late 2026, initially via select partners). The much-cited ~$20,000 figure is an analyst target, not a Figure price, and it would not include whatever recurring software, connectivity, or service costs a deployed home humanoid might carry. Treating it as a real price today would be guessing.

Want to model what a humanoid deployment might cost? There’s no published Figure rate to plug in, but you can sketch scenarios with the Figure pricing calculator, and compare how other physical-AI peers structure their meters across the pricing blueprint — Figure itself publishes no rate of its own.

Pricing evolution : Figure pricing history and changes

Figure has never had a public price to change. Its “pricing evolution” is really a commercial-posture evolution: funded humanoid startup, then OpenAI partner, then fully in-house AI builder and vertical manufacturer — all while keeping its actual commercial offering behind sales-led, milestone-based pilots. The milestones below are reconstructed from primary announcements and a live 2026-06-14 site check.

Cadence

Tracked range: 2024 Q1–2026 Q2. Zero public price changes across the company’s life — there has never been a published price to revise. Quarters not listed had no relevant public pricing event.

Notable changes

• 2024-02-29 — $675M Series B at a $2.6B valuation, alongside an OpenAI collaboration agreement; backers include Microsoft, NVIDIA, Jeff Bezos, Intel Capital and ARK Invest. Commercializing humanoids, but no price card (PR Newswire).
• 2025-02 — Drops OpenAI for in-house models and unveils the Helix Vision-Language-Action model — a capability bet, not a priced SKU (figure.ai/news/helix).
• 2025-03-15 — BotQ in-house manufacturing facility announced, stated capacity up to 12,000 humanoids/year; no per-unit price (figure.ai/news/botq).
• 2025-09-16 — Series C exceeding $1B at a $39B post-money valuation, led by Parkway Venture Capital; funds target BotQ scale-up and deployments (figure.ai/news/series-c).
• 2026-06-14 — Live check confirms no public pricing: figure.ai/pricing 404s; the only CTAs are a newsletter sign-up and a contact form.

What’s unique : Figure’s distinctive pricing mechanics

1. A $39B valuation with no price tag. Figure is valued at $39B post-money — more than many public software companies — yet publishes no price for the humanoid robots that the valuation is built on. The entire monetized surface sits behind a sales conversation and milestone-based contracts. For a company whose product is a physical robot deployed into bespoke industrial environments, opacity is the deliberate packaging: each deployment is engineered and quoted, so there is no list price to commoditize against.

2. Milestone-based, not metered. Where the rest of this corpus is split between published meters (Waymo per ride) and RaaS rates (Agility Robotics per robot-hour), Figure’s known commercial structure — the BMW agreement — is described as milestone-based. Value is tied to hitting deployment milestones inside a real production line, not to a per-hour or per-unit meter the buyer can read in advance. That’s an outcome-shaped contract, and it resists a public unit price by design. (See outcome-based pricing trends.)

3. Owning the whole stack changes what’s even priceable. By building Helix in-house and manufacturing via BotQ, Figure controls the model, the hardware, and the production line. That vertical integration means there is no third-party rate to pass through and no component price to anchor on — the “value metric” a buyer underwrites is a robot that does the job, an outcome that doesn’t decompose neatly into a published unit. For how to pick a metric when the deliverable is an outcome rather than a unit, see choosing the right usage metric.

Strengths & weaknesses

Billing UX : Figure billing controls and transparency

• Billing controls — None are public. There is no self-serve dashboard, no usage meter, and no plan-management UI on the site; commercial terms are handled entirely through a sales relationship and a custom contract.
• Usage visibility — Not applicable to a public buyer. Any operational telemetry (robot uptime, tasks completed) would live in whatever fleet-management tooling Figure provides to a deployment customer, not in a public console.
• Payment options — Not disclosed. Enterprise deployments are presumably invoiced under custom, milestone-based contracts negotiated through the contact form; no card checkout or public billing portal exists.
• Transparency — Deliberately low on the commercial side. Figure is communicative about technology (Helix, BotQ, demo videos, a public “Master Plan”) and silent about price — the inverse of a self-serve SaaS company.

Strategic wins : Why Figure’s pricing decisions worked

1. A marquee pilot instead of a price sheet

By landing BMW Manufacturing as a milestone-based reference deployment, Figure proved its robots in the hardest possible setting — a live automotive line — without ever publishing a rate card. The pilot does the selling that a price page can’t: it shows a Fortune-tier operator trusting Figure with real production, which is far more persuasive to the next enterprise buyer than a number. This is the physical-AI version of selling outcomes over units, a pattern echoed across how AI companies are rethinking what they meter.

2. No list price, no commoditization

The humanoid race is crowded and fast. By keeping every deal bespoke and milestone-based, Figure avoids a public rate that rivals could undercut line-by-line and keeps pricing tied to the engineered value of a working deployment. For a robot that has to be integrated into each customer’s environment, a list price would understate the real (services-heavy) cost and invite a race to the bottom — quoting protects both margin and positioning. See usage-based pricing strategy for when metering the outcome beats metering the unit.

3. Capital as the substitute for revenue transparency

With a $39B valuation and backers like NVIDIA, Microsoft and Bezos, Figure can fund a long, capital-intensive build (in-house AI plus a 12,000-unit/year factory) without needing a self-serve revenue engine yet. That war chest lets it stay disciplined about not publishing a price until the product and unit economics are ready — selling on capability and trust in the meantime. Related: outcome-based pricing trends.

Areas to improve : Gaps in Figure’s pricing approach

1. Publish something — even the model, if not the number

A fully dark price page (a literal 404) maximizes evaluation friction. Even a one-line description of how deployments are priced — purchase vs lease vs RaaS, what’s bundled with Helix, whether there’s a per-robot-hour component — would let operators self-qualify instead of cold-emailing. The total opacity invites exactly the cost-unpredictability anxiety that slows enterprise buyers.

2. Get ahead of the ~$20,000 narrative

A specific consumer price (”~$20,000”) is circulating widely with no official source. That vacuum lets third parties define Figure’s price story for it. A clear statement — even “Figure 03 is not yet for sale; pricing will be announced when it is” — would replace speculation with a credible anchor and protect the eventual launch from being judged against a number Figure never set.

3. Clarify the RaaS-vs-purchase question

Operators evaluating humanoids think first about robots-as-a-service vs capex. Figure says nothing publicly about which it offers. Surfacing even a high-level commercial structure (as RaaS peers like Agility Robotics do) would shorten the evaluation cycle and reduce reliance on a cold sales conversation for every interested buyer. Compare how other AI companies stage enterprise transparency.

Key takeaways

1. No public price is itself a pricing decision. Figure publishes zero commercial pricing and routes everything through sales and milestone-based contracts — a deliberate posture for a bespoke, integration-heavy physical product, not an oversight. Opacity protects margin and negotiating leverage in a crowded humanoid race.
2. Funding numbers are not product prices. Figure’s only public figures — a $675M Series B at $2.6B and a Series C over $1B at $39B — describe what investors paid for equity, not what a customer pays for a robot. Conflating the two is the easiest mistake to make here.
3. A reference deployment can replace a rate card. The BMW Spartanburg pilot (30,000+ vehicles assisted over ~11 months) does the persuasion a price page would, while keeping every deal bespoke.
4. Owning the stack makes the product harder to unit-price. With Helix in-house and BotQ manufacturing, the deliverable is “a robot that does the job” — an outcome that resists decomposition into a published per-unit meter.
5. The consumer price is a rumor, not a fact. The ~$20,000 Figure 03 figure is third-party speculation; treat any “Figure costs $X” claim with suspicion until Figure itself publishes one.

UBP implications

1. Physical, integration-heavy products often price by outcome, not unit. When a deployment requires bespoke engineering into each customer’s environment, a public per-unit rate understates the real cost and commoditizes the offering. Figure’s milestone-based contracts show UBP practitioners how to tie payment to delivered outcomes when the “unit” isn’t fungible.
2. A meter only helps when the thing metered is comparable. Waymo can publish a per-ride fare and a RaaS vendor a per-robot-hour rate because those units are comparable across trips or shifts. Figure’s deployments aren’t — so it quotes. The lesson: match transparency to fungibility; publish meters for commodities, quote bespoke outcomes.
3. Beware pricing the rumor. When a company stays silent, the market invents a number (here, ~$20,000). UBP designers should recognize that not setting a public anchor cedes the narrative — and decide deliberately whether the credibility cost of an information vacuum outweighs the margin protection of opacity.

Sources

• Figure — official site (no pricing page; newsletter + contact CTA) (accessed 2026-06-14)
• Figure — /pricing returns HTTP 404 (accessed 2026-06-14)
• Figure Exceeds $1B in Series C Funding at $39B Post-Money Valuation (PR Newswire / figure.ai) (accessed 2026-06-14)
• Figure raises $675M at $2.6B valuation, signs OpenAI collaboration (PR Newswire) (accessed 2026-06-14)
• Helix: A Vision-Language-Action Model for Generalist Humanoid Control (figure.ai) (accessed 2026-06-14)
• BotQ: A High-Volume Manufacturing Facility for Humanoid Robots (figure.ai) (accessed 2026-06-14)
• Figure announces commercial agreement with BMW Manufacturing (PR Newswire) (accessed 2026-06-14)
• Figure reaches $39B valuation in latest funding round (TechCrunch) (accessed 2026-06-14)
• Browse the pricing blueprint corpus

Bottom line

Figure is a clear sales-only case in the embodied-AI corpus: a Brett Adcock-founded humanoid company, $39B post-money after a Series C exceeding $1B, that publishes no price for anything it makes. Its humanoids (Figure 03), its in-house Helix AI, and its BotQ-built hardware all reach customers through sales-led, milestone-based enterprise pilots — figure.ai/pricing is a literal 404 — with BMW Manufacturing as the flagship reference. The only public numbers are funding and valuation, which measure investor conviction, not customer cost; the widely-cited ~$20,000 consumer figure is third-party speculation Figure has never confirmed. The bet: own the whole stack, prove it in marquee deployments, and keep the price behind a sales conversation until the unit economics are ready to show.

Want to compare Figure against physical-AI peers? See the per-ride meter of Waymo and the robots-as-a-service model of Agility Robotics, or browse the full pricing blueprint.