# BE Quarter forecast · Q3 2026 > Bloom Energy reported Q3 2026. These are the numbers we published before the release, with a cutoff of 16 October 2026, 13:10 ET. Canonical: https://pricedin.markets/be/fq3-2026/preview/ Thursday 22 October, after the close · reported Bloom Energy reported Q3 2026. These are the numbers we published before the release, with a cutoff of 16 October 2026, 13:10 ET. [Release review →](https://pricedin.markets/be/fq3-2026/result/) BE · earnings preview · our forecast as of 2026-10-16 13:10:00 ET · 6 AI forecasters · 80% ranges, 1 in 10 to 9 in 10 ## BE reports Q3 2026 on 22 October 2026 release after the close · call 17:00 ET Next report · Q3 2026 Sales $1.10B +113% on Q3 2025 Adjusted EPS $0.76 +406% on Q3 2025 Quarter after · Q4 2026 Sales $1.19B +53% on Q4 2025 Adjusted EPS $0.83 +85% on Q4 2025 ### Q3 2026 · the quarter BE reports ##### Sales PricedIn **$1.10B** $0.91B – $1.32B $0.91B $1.10B $1.32B ##### adjusted EPS PricedIn **$0.76** $0.52 – $0.98 $0.52 $0.76 $0.98 **Middle** the forecast **80% range** 1 in 10 chance below this bar, 1 in 10 above it. These are the low and high printed by each number. All six put sales at about $1.1B; they differ on profit. Gemini and Muse Wonder are lowest at $0.72 a share; Gemini keeps margin at the company's full-year level, "rather than extrapolating a higher underlying jump from the second quarter." Sol is highest at $0.85, with "a modest delivery ramp rather than the old large December seasonal spike." Sol's range is also widest, $850M to $1.5B, because one big buyer "leaves substantial quarterly delivery risk." Third-quarter sales hold near $1.1B, level with the record second quarter, because big campus deliveries follow customers' site readiness. Sales **$1.10B** $0.91B – $1.32B · +113% on Q3 2025 ($0.52B) Adjusted EPS **$0.76** $0.52 – $0.98 · +406% on Q3 2025 ($0.15) #### Investment and cash generation Base case Capital spending **$59.2M** Free cash flow **$92.5M** Free cash flow: operating cash flow less capital spending. Bloom Energy makes fuel-cell power systems, now sold mostly to AI data centers that need power fast. In the second quarter of 2026 sales passed $1B for the first time, reaching $1.1B, with gross margin of 33.4% and adjusted earnings of $0.78 a share. The company raised its full-year sales outlook to $3.9B to $4.2B. The company warns that quarters are lumpy: one or two customers can lead revenue in any quarter, and one buyer took 73% of second-quarter revenue. It says customers' site readiness, not its factory, sets the pace, so a single slipped campus can move a quarter by hundreds of millions. All six models put third-quarter sales at about $1.1B, about 3.6% above the second quarter on average. Adjusted earnings average $0.76 a share, a little below the $0.78 just reported, because the $37.4M tariff refund that lifted second-quarter margin is not expected to repeat and operating costs keep rising in dollars. The ranges are wide. On average the low is $910M and the high $1.3B. Sol spans $850M to $1.5B, the widest, and Muse Wonder has the lowest low at $800M. Earnings per share come from sales near $1.1B at a gross margin of about 33%, adjusted operating costs rising slowly, almost no tax thanks to past losses, and a slowly rising share count. #### Sales by quarter Q1 2025 $0.33B +39% Q2 2025 $0.40B +20% Q3 2025 $0.52B +57% Q4 2025 $0.78B +36% Q1 2026 $0.75B +130% Q2 2026 $1.07B +166% Q3 2026 $1.10B +113% Each percent is on the same quarter a year earlier. The last row is the models' forecast. #### The quarter forecast, line by line | Line | Q2 2026 reported | Q3 2025 reported | Q3 2026 forecast | Q4 2025 reported | Q4 2026 forecast | | --- | --- | --- | --- | --- | --- | | Total revenue | $1.07B | $0.52B | $1.10B | $0.78B | $1.19B | | Sales growth, year on year | +166% | +57% | +113% | +36% | +53% | | Net income (loss) attributable to common stockholders | $196M | −$23M | $187M | $1M | $211M | | Diluted (in shares) | 0.32B | 0.23B | 0.33B | 0.26B | 0.33B | | Adjusted EPS | $0.78 | $0.15 | $0.76 | $0.45 | $0.83 | The lines as the company's statement shows them. The forecast columns are the average of the models, line by line. Sales growth is the change on the same quarter a year earlier. Earnings per share is the headline's: the company's own for the reported quarter, and for a forecast quarter worked out from the averaged lines with the company's adjustments. #### Biggest buyers' share of revenue One customer, second quarter 2026 73% Largest customer, first half 2026 44% Related-party customer, first half 2026 21% **Where each number comes from** **One customer, second quarter 2026** “During the three months ended June 30, 2026, revenue from one customer*, which is not our related party, accounted for approximately 73% of our total revenue.” **Largest customer, first half 2026** “accounted for approximately 44% and 21% of our total revenue” **Related-party customer, first half 2026** “accounted for approximately 44% and 21% of our total revenue” A few large buyers take most sales, so one slipped campus can move a quarter. #### What it rests on **Second-quarter sales** $1.1B, up 165.5% on a year earlier “Revenue of $1,065.4 million increased 165.5% compared to $401.2 million.” Second-quarter 2026 earnings release **Biggest buyer's share** One customer took 73% of second-quarter revenue “During the three months ended June 30, 2026, revenue from one customer*, which is not our related party, accounted for approximately 73% of our total revenue.” Amended second-quarter 2026 quarterly report **One-off tariff refund** A $37.4M refund of past import tariffs lifted second-quarter product profit “the recognition of a $37.4 million recovery of previously paid import tariffs” Second-quarter 2026 quarterly report **Deliveries are lumpy** One or two customers can lead revenue in any quarter “large campus deliveries are lumpy. One or two customers could lead our revenue in any given quarter” Second-quarter 2026 earnings call **Full-year gross margin outlook** About 34% on the company's adjusted basis “•Non-GAAP Gross Margin: ~34%” Second-quarter 2026 earnings release #### Low and high Base the middle forecast Key assumptions - ****Sales of about $1.1B, a small step up from the second quarter.** Signed deliveries plus same-year orders sit inside the raised outlook; customers' site dates set the timing. *6 of 6*** - **Opus 5.5** I put the third quarter at $1.10 billion (a small step up, since deliveries are lumpy and the company says timing follows customer readiness) - **Grok 4.7** Signed deliveries plus same-year orders already sit inside the raised full-year range. - **GLM 5.3** giving a third quarter near $1.10 billion - **Gemini 4 Argon** 1.08 billion dollars in the third quarter - **GPT-6.1 Sol** My third-quarter sales middle is $1.15 billion - **Muse Wonder** split between the third and fourth quarters - ****Gross margin of about 33%, with no repeat of the $37.4M tariff refund.** The refund was one-off, and new import tariffs took effect in late July. *4 of 6*** - **Gemini 4 Argon** Because the second-quarter tariff refund was non-recurring and new tariffs took effect in the third quarter - **Muse Wonder** About thirty seven million dollars of tariff recovery flattered first half product profit and will not repeat at the same size - **GPT-6.1 Sol** I assume no repeat exceptional recovery or new exceptional warranty charge in the middle - **Opus 5.5** On balance I kept the base margin near the company's 34% to 35%. - ****Operating costs rise in dollars but far slower than sales, keeping operating margin near 17%.** Second-quarter sales grew 166% while operating expenses grew 48%. *3 of 6*** - **Grok 4.7** Sales grew 166 percent in the second quarter while operating expenses grew 48 percent - **GPT-6.1 Sol** I forecast adjusted operating expenses of $135 million and $145 million in the next two quarters - **Opus 5.5** adjusted operating costs rising from $126 million to about $137 to $145 million a quarter - ****Tax stays near zero and the diluted share count rises slowly.** Federal loss carryforwards of $2.3B shelter profit. *3 of 6*** - **Opus 5.5** $2.3 billion of federal losses, plus large stock compensation deductions, should keep the tax charge small through 2027. - **GLM 5.3** with diluted shares growing about 4 million per quarter - **GPT-6.1 Sol** Dividing by 328 million and 330 million diluted shares gives the quarterly earnings in the form. If all of this happens **$1.10B** sales **$0.76** adjusted EPS **$59.2M** Capital spending **$92.5M** Free cash flow Cash walk not supplied by every model. Low about 1 chance in 10 of worse What the low case assumes - ****One or two large campus deliveries slip into later quarters.** One buyer took 73% of second-quarter revenue, so one site moves the quarter. *6 of 6*** - **Opus 5.5** delays in customer site readiness for large campus deliveries - **GPT-6.1 Sol** Large delivery windows move into later quarters or years - **Grok 4.7** One customer was about three quarters of first-half sales, so a pause at that campus is hard to replace in a quarter. - **Muse Wonder** Customer site readiness and delivery timing slips - **Gemini 4 Argon** Customer site readiness and greenfield construction delays pushing large campus deliveries into later periods - **GLM 5.3** so one or two delayed sites can dent a quarter or a year - ****Expediting, warranty and tariff costs pull gross margin below about 33%.** New import tariffs of 10% to 12.5% began in late July; a $58.3M warranty reserve hit the first half. *4 of 6*** - **Gemini 4 Argon** Import tariff increases of ten to twelve and a half percent and elevated expediting costs pressuring product gross margins - **GPT-6.1 Sol** Warranty and input costs weaken gross profit - **Muse Wonder** Margin pressure from expediting, warranty and input costs - **GLM 5.3** Expediting costs and service-maintenance timing compress gross margin below the roughly 34% outlook - ****Same-year quick-power orders convert more slowly.** The outlook counts on booking and shipping some orders within the same year. *3 of 6*** - **Opus 5.5** slower conversion of in-year bookings - **Muse Wonder** Slower in-year bookings for quick power - **GPT-6.1 Sol** Financing and customer site readiness slow new orders - ****Higher interest rates make buyers' project financing dearer.** The 10-year Treasury yield rose to about 5.2% by mid-October. *2 of 6*** - **Muse Wonder** Higher long-term interest rates raising project financing costs - **Opus 5.5** financing costs for third-party buyers If all of this happens **$0.91B** sales · −18% vs base **$0.52** EPS · −31% vs base Sales would fall to about $910M on average, 15% below the second quarter, with earnings near $0.52 a share. One or two large campuses would have to slip past September 30 without same-year orders filling the gap, while expediting, warranty and tariff costs press margin. Muse Wonder goes furthest, to $800M and $0.40, because it adds slower same-year quick-power orders and higher long-term interest rates raising project financing costs to the site delays the others share; most others stop between $920M and $980M. Quarters have swung hard before: sales fell 43% from the end of 2024 into the first quarter of 2025, and dipped 3.4% at the start of 2026. The six lows run from $0.80B (Muse Wonder) to $0.98B (GLM 5.3). First sign Third-quarter results, expected in late October, showing sales below $1B or talk of customers rescheduling deliveries. High about 1 chance in 10 of better What the high case assumes - ****Customer sites finish early and big deliveries pull into the third quarter.** Management says site readiness, not the factory, sets the pace of growth. *3 of 6*** - **GPT-6.1 Sol** Rapid site completion and redeployment minimize delivery delays - **Muse Wonder** deliveries pull forward - **Grok 4.7** customer sites take the power - ****Same-year bookings from cloud providers keep outrunning plan.** Backlog is growing faster than revenue, and all major US cloud providers have approved Bloom's systems. *3 of 6*** - **GLM 5.3** In-year bookings keep outpacing plan as time-to-power customers book and convert inside the same fiscal year - **Opus 5.5** more in-year bookings from hyperscalers and new cloud providers - **GPT-6.1 Sol** Multiple large cloud customers turn approvals into repeated equipment orders - ****Factory output keeps pace, with capacity added in steps each quarter.** The plant moves to 2 gigawatts by year end and can reach 5 gigawatts. *3 of 6*** - **Grok 4.7** Capacity is now added in hundreds of megawatts a quarter, not in one annual lump. - **Opus 5.5** capacity added well beyond 2 gigawatts - **GPT-6.1 Sol** Factory additions and suppliers scale without quality problems - ****Pricing for speed holds gross margin up despite faster shipments.** Customers value time to power most and pay for quick delivery. *3 of 6*** - **Opus 5.5** pricing power as customers pay for speed - **Muse Wonder** Stronger pricing for time to power and continued product cost reduction - **GPT-6.1 Sol** Pricing for quick power and cost reductions widen margins If all of this happens **$1.32B** sales · +19% vs base **$0.98** EPS · +29% vs base Sales would reach about $1.3B on average, 24% above the second quarter, with earnings near $0.98 a share. Customers would have to finish sites early, same-year orders keep arriving, and big deliveries pull forward into September, with pricing for speed holding margin. Sol goes furthest, to $1.5B and $1.20, because it holds that several large cloud customers "turn approvals into repeated equipment orders" and finish sites quickly; the others stop between $1.2B and $1.4B. The second quarter itself jumped 42% on the first, so a quarter far above plan has just happened. The six highs run from $1.20B (Gemini 4 Argon) to $1.48B (GPT-6.1 Sol). First sign Third-quarter sales above about $1.25B at the late-October report, with another raise of the full-year outlook. #### Watch The third-quarter report, expected in late October: sales against $1.1B, gross margin against about 33%, and whether the full-year range moves. A raise above $4.2B points higher; customer rescheduling, slower deposits or a new warranty charge point lower. ### Q4 2026 · the quarter after ##### Sales PricedIn **$1.19B** $0.98B – $1.43B $0.98B $1.19B $1.43B ##### adjusted EPS PricedIn **$0.83** $0.59 – $1.08 $0.59 $0.83 $1.08 **Middle** the forecast **80% range** 1 in 10 chance below this bar, 1 in 10 above it. These are the low and high printed by each number. Middles sit close, from $1.1B to $1.3B. Muse Wonder is lowest, keeping the year at "sales near the midpoint and adjusted earnings per share near the middle of the range, split between the third and fourth quarters." Sol is highest at $1.3B, putting the year "slightly above the sales ceiling because new orders are converting quickly." Sol's range is also widest, $950M to $1.6B. Fourth-quarter sales rise to about $1.2B as deliveries ramp steadily, without the old year-end spike. Sales **$1.19B** $0.98B – $1.43B · +53% on Q4 2025 ($0.78B) Adjusted EPS **$0.83** $0.59 – $1.08 · +85% on Q4 2025 ($0.45) #### Investment and cash generation Capital spending: not supplied by every model Free cash flow: not supplied by every model Free cash flow: operating cash flow less capital spending. The fourth quarter used to be Bloom's biggest by far: sales jumped 73% into the end of 2024 and 50% into the end of 2025. The company now says that pattern is gone, and that what was once seasonality is simply delivery timing based on customer readiness. The raised outlook of $3.9B to $4.2B, less first-half sales of $1.8B, leaves the second half to deliver the rest. The models' full-year view leaves about $1.2B for the fourth quarter. On average the models expect $1.2B, up 7.9% on their third quarter, and adjusted earnings of $0.83 a share. Muse Wonder is lowest at $1.1B; Sol is highest at $1.3B and $0.97, the only model placing the year above the top of the company's range. Ranges run from about $982M to $1.4B on average. Margins should edge up as the factory doubles to 2 gigawatts by year end and unit costs fall, though new import tariffs and the Oracle share award push the other way. Fourth-quarter earnings per share build on sales near $1.2B, gross margin around 33% to 34%, operating costs rising more slowly than sales, almost no tax, and a share count that keeps creeping up. #### What it rests on **Seasonality gone** Year-end spikes have given way to delivery timing set by customer readiness “What was once seasonality is now simply delivery timing based on customer readiness.” Second-quarter 2026 earnings call **Full-year sales outlook** $3.9B to $4.2B, raised in July “•Revenue: $3.9B - $4.2B” Second-quarter 2026 earnings release **Capacity doubling** Factory to reach 2 gigawatts a year by the end of 2026 “We plan to double our factory capacity from 1 gigawatt to 2 gigawatts by the end of 2026.” 2025 annual report **Oracle share award** Share-based consideration reduces revenue as Oracle systems are delivered “is amortized as a reduction of revenue as the underlying Energy Server systems sold under the Oracle arrangement are delivered.” Second-quarter 2026 quarterly report #### Low and high Base the middle forecast Key assumptions - ****Sales of about $1.2B, a steady step up rather than a December spike.** The company says seasonality is now delivery timing based on customer readiness. *5 of 6*** - **GPT-6.1 Sol** I use a modest delivery ramp rather than the old large December seasonal spike. - **Grok 4.7** it set the quarterly split of the middle rather than a year-end spike - **GLM 5.3** I split that roughly evenly with a modest fourth-quarter skew - **Gemini 4 Argon** 1.18 billion dollars in the fourth quarter - **Opus 5.5** the fourth at $1.20 billion - ****The full year lands a little above the middle of the $3.9B to $4.2B range.** The company has raised its outlook after each quarter this year. *3 of 6*** - **Opus 5.5** I put the full year a little above the midpoint - **Grok 4.7** I put this year's sales a little above that middle - **GLM 5.3** just above the guidance midpoint because the company has beaten its own outlook every quarter this year - ****Operating costs rise in dollars but much more slowly than sales.** Research and administrative costs are largely fixed against fast-growing sales. *3 of 6*** - **GPT-6.1 Sol** I forecast adjusted operating expenses of $135 million and $145 million in the next two quarters - **Opus 5.5** operating costs growing well below revenue - **Grok 4.7** Overhead grows much more slowly than sales, so profit rises faster than revenue. - ****The Oracle share award trims sales as systems ship, already inside the outlook.** The award is amortized against revenue as Oracle systems are delivered. *2 of 6*** - **GPT-6.1 Sol** I allow roughly $40 million to $50 million of additional second-half amortization within forecast sales and margins - **Grok 4.7** This year's guide already nets whatever portion management expects to deliver this year If all of this happens **$1.19B** sales **$0.83** adjusted EPS Low about 1 chance in 10 of worse What the low case assumes - ****Year-end campus deliveries slip into 2027.** A few buyers take most sales, and site readiness sets delivery dates. *4 of 6*** - **Opus 5.5** delays in customer site readiness for large campus deliveries - **Muse Wonder** Customer site readiness and delivery timing slips - **Gemini 4 Argon** Customer site readiness and greenfield construction delays pushing large campus deliveries into later periods - **GPT-6.1 Sol** Large delivery windows move into later quarters or years - ****Financing for buyers slows as interest rates rise.** The 10-year Treasury yield is about 5.2%, and most deals rely on third-party financing. *3 of 6*** - **Muse Wonder** Higher long-term interest rates raising project financing costs - **Opus 5.5** financing costs for third-party buyers - **GPT-6.1 Sol** Financing and customer site readiness slow new orders - ****Warranty, tariff and expediting costs press gross margin.** New import tariffs began in late July and warranty reserves rose in the first half. *3 of 6*** - **Gemini 4 Argon** Import tariff increases of ten to twelve and a half percent and elevated expediting costs pressuring product gross margins - **GPT-6.1 Sol** Warranty and input costs weaken gross profit - **Muse Wonder** Margin pressure from expediting, warranty and input costs - ****Faster Oracle deliveries cut reported sales through the share award.** The remaining award reduces revenue as Oracle systems ship. *2 of 6*** - **Grok 4.7** Share consideration of about 300 million dollars reduces reported sales as those systems ship. - **GPT-6.1 Sol** A faster Oracle delivery mix could reduce reported sales more than my base assumes. If all of this happens **$0.98B** sales · −18% vs base **$0.59** EPS · −30% vs base Sales would fall to about $982M on average, 11% below the third quarter, with earnings near $0.59. Year-end campuses would have to slip into 2027 and same-year orders stall, leaving the year near the bottom of the outlook, while faster Oracle deliveries cut reported sales through the share award. Muse Wonder goes furthest, to $900M and $0.50, because it adds higher long-term interest rates raising project financing costs and a full-year miss below the raised outlook; most others stop at about $1.0B. The whole of 2026 rests on the second half: first-half sales were $1.8B against an outlook of $3.9B to $4.2B. The six lows run from $0.90B (Muse Wonder) to $1.04B (Gemini 4 Argon). First sign A late-October third-quarter report that holds the full-year range or trims it toward $3.9B. High about 1 chance in 10 of better What the high case assumes - ****Same-year bookings keep outrunning plan.** The company reserves factory capacity for quick-power customers who book and take delivery within the year. *3 of 6*** - **GLM 5.3** In-year bookings keep outpacing plan as time-to-power customers book and convert inside the same fiscal year - **Opus 5.5** more in-year bookings from hyperscalers and new cloud providers - **GPT-6.1 Sol** Multiple large cloud customers turn approvals into repeated equipment orders - ****Financing partners draw their commitments quickly.** Brookfield raised its commitment to $25B, and other lenders added $2.6B. *3 of 6*** - **Opus 5.5** large financing commitments drawn against real contracts - **Muse Wonder** Financing partners expanding commitments quickly - **Grok 4.7** Faster drawdown of that shelf is a driver of the high case. - ****Pricing for speed lifts margins.** Customers pay for power delivered in months instead of years. *3 of 6*** - **Opus 5.5** pricing power as customers pay for speed - **Muse Wonder** Stronger pricing for time to power and continued product cost reduction - **GPT-6.1 Sol** Pricing for quick power and cost reductions widen margins - ****The expanded factory runs full as new lines come on.** Capacity reaches 2 gigawatts by year end, with room for 5 gigawatts. *2 of 6*** - **Gemini 4 Argon** Rapid utilization of existing five gigawatt annual manufacturing capacity and addition of new factory capacity with payback periods of only a few months - **GPT-6.1 Sol** Factory additions and suppliers scale without quality problems If all of this happens **$1.43B** sales · +20% vs base **$1.08** EPS · +30% vs base Sales would reach about $1.4B on average, 29% above the third quarter, with earnings near $1.08. Same-year bookings would have to keep outrunning plan, the expanded factory ship at full pace into December, and financing partners draw their commitments fast. Sol goes furthest, to $1.6B and $1.40, because it alone expects large repeat fleet orders from several cloud customers with sites completed quickly; the others stop between $1.3B and $1.4B. The company has raised its full-year sales outlook twice this year, from $3.1B to $3.3B in February to $3.9B to $4.2B in July. The six highs run from $1.33B (Gemini 4 Argon) to $1.65B (GPT-6.1 Sol). First sign Another raise of the 2026 outlook, above $4.3B, at the late-October third-quarter report. #### Watch The late-October third-quarter report and its full-year range: a raise past $4.2B implies a stronger fourth quarter, a hold implies about $1.2B. Also watch customer deposits, the pace of Oracle deliveries, and comments on new tariffs or warranty costs.