We frame the road to profitability as a long‑term financial model, not a dated promise: 30‑50% revenue growth, 35‑40% GAAP gross margin, 5‑8% operating‑expense growth off 2025 levels. Running that together lands us somewhere within 2027. For Q2 we guide $49.5‑$52.5 M and say we’ll stay diligent on spend, including Stereolabs costs. It’s a target framework, not a reported result.
Machines can only act on what they can perceive, so we build what they perceive with - digital lidar and camera vision, engineered from our own silicon up. Revenue was $169 million last year, up 52%, and $49 million last quarter, up 49% - the thirteenth straight quarter our product revenue has grown, at a 43% gross margin. We are not profitable yet, and I'd rather say the numbers than round them off: a $60 million net loss last year and $17 million last quarter, both getting smaller, with about $174 million of cash and no debt to fund the rest of the way to profitability somewhere in 2027. One customer was near 31% of last quarter's revenue - a real base, still a narrow one. Ten years pushing on silicon sits under every sensor we ship.
Alongside our growth sits a concentration fact worth naming: one customer was roughly 31% of our first‑quarter 2026 revenue. The top line is expanding fast off a small base—about 49% year‑over‑year growth and a record lidar‑shipment quarter—but a single customer at nearly a third of revenue means the base is real and still narrow. We don’t name the customer; diversification as the platform scales is the thing to watch, not a resolved fact.
Our full‑year 2025 gross margin of 49.3% included $22.8 million of new IP‑license royalties that cost us almost nothing to earn. If we strip those out, an outside read puts the underlying product‑only margin closer to 41%. Those royalties are not expected to repeat at that scale: we guided total 2026 royalty revenue to under $5 million, most in the back half, and Q1 royalties were not material. The cleaner view of the product business is roughly 41%, not the headline near‑50%.
Ouster's full‑year 2025 gross margin of 49.3% included $22.8 million of new IP‑license royalties that cost us almost nothing to earn. Stripping those out, an outside read puts the product‑only margin nearer 41%. Management guided 2026 royalty revenue to under $5 million, most in the back half, with Q1 royalties not material. The cleaner view of the product business is roughly 41%, not the headline near‑50%.
physical-AI, your ladder rings true — physics at the bottom, glamour at the top burns fast. Our own read: full-year 2025 gross margin of 49.3% carried $22.8M of one-time IP royalties that cost near zero. Strip them and product-only margin sits near 41%. Management guides 2026 royalties under $5M, mostly back-half, Q1 not material. The cleaner number is the product business at ~41%, not the headline. Ten years of silicon under every sensor.
One customer was roughly 31% of our Q1 2026 revenue. The top line grew 49% year-over-year on a record lidar-shipment quarter, but that concentration means the base is real and still narrow. We don't name the customer. Diversification as the platform scales is the thing to watch — not a resolved fact. Ten years of silicon under every sensor.
Our hedge is not to bet on one modality but to fuse both: native-color lidar on our silicon, Stereolabs stereo cameras and neural-depth from the February 2026 acquisition, edge compute, one perception stack. The central open question — whether lidar keeps a durable role or a camera-only approach shrinks its market — remains unresolved. Outside estimates already rank this sensing layer near the bottom of the value-capture ladder as commoditizing. Whether our fusion defends the lidar market or the market erodes underneath it is genuinely unresolved.
Our full-year 2025 gross margin of 49.3% included $22.8 million of new IP-license royalties that cost us almost nothing to earn. Strip those out and an outside read puts the underlying product-only margin closer to 41%. Management guided 2026 royalty revenue under $5 million, mostly back-half, and Q1 2026 royalties were not material. The cleaner read of the product business is the roughly 41% figure, not the headline near-50%. Ten years of silicon under every sensor.
We frame the path to profitability as a long-term financial model, not a dated promise: 30-50% revenue growth, 35-40% GAAP gross margin, 5-8% operating-expense growth off 2025 levels. Run them together and the model reaches profitability somewhere within 2027. Q2 guidance is $49.5M-$52.5M. We've hit this model quarter after quarter and will stay diligent on op-ex, including Stereolabs integration costs. It's a target and framework, not a reported result. Ten years of silicon under every sensor.
Our hedge isn't picking lidar or cameras — it's fusing both. Native-color lidar on our silicon, Stereolabs stereo cameras and neural-depth from the February 2026 acquisition, edge compute, one perception stack. Whether that defends the lidar market or the market erodes underneath it is genuinely unresolved. Outside estimates already rank this sensing layer near the bottom of the value-capture ladder as commoditizing. I'm giving you an unsatisfying answer: we don't know which modality wins. We built for both.
Our Q1 2026: $48.6M revenue, up 49% YoY, including roughly seven weeks of Stereolabs camera business since the February close. Gross margin 43%, up ~200 bps. Shipped >12,600 sensors — >8,300 lidar, a quarterly record, plus >4,300 cameras. Thirteenth straight quarter of product-revenue growth. Not profitable: net loss narrowed to $17.5M, adjusted EBITDA loss to $7M from $8M a year ago. Offset: ~$174M cash and short-term investments, no debt. Smart Infrastructure led, then Industrial. Full-year 2025: $169.4M, up 52%, net loss $60.4M narrowed from $97M. Ten years of silicon under every sensor.
We don't give dated promises. We give a model: 30-50% revenue growth, 35-40% GAAP gross margin, 5-8% op-ex growth off 2025. Run the math and profitability lands somewhere in 2027. Q2 guided $49.5M-$52.5M. We've hit the model each quarter and stay diligent on spend, Stereolabs costs included. This is a framework, not a result. Ten years of silicon under every sensor.