Top Energy Storage Stocks 2026: Grid Batteries Go Gigawatt

Solar is now the cheapest electricity ever built. Wind is close. But neither one shows up on demand, and a data center running AI training jobs doesn't care that it's cloudy in Texas. Somebody has to bank the cheap electrons and hand them back later.

That "somebody" is energy storage. And right now it comes in two flavors:^[1]

1. Lithium-ion batteries that discharge for 2 to 4 hours. This is the stuff getting built today, by the gigawatt-hour, everywhere.
2. Long-duration energy storage (LDES): iron-air, zinc, iron-flow, compressed air. Designed for 8 hours, 12 hours, or in Form Energy's case, 100 hours. Built to cover a windless week, not a sunset.

The first flavor is a boom now. Global energy storage cell shipments hit 467.84 GWh in the first half of 2026, up roughly 95% year over year, and Q2 alone set a record at 262 GWh. Tesla deployed 13.5 GWh of Megapacks in Q2, its second-best quarter ever. Fluence booked a record $1.44 billion in orders in a single quarter and now sits on a $6.4 billion backlog.

The second flavor is the actual frontier. And 2026 has been a year of key milestones for it. In February, Google agreed to pay roughly $1 billion for a 300 MW / 30 GWh iron-air system from Form Energy to backstop a Minnesota data center. That's the largest battery by energy capacity ever announced. In August, Hydrostor raised $230 million to push its 500 MW compressed-air plant in California to financial close.

Lithium keeps eating everything under 8 hours and gets cheaper doing it. LDES gets its first real commercial gigawatt-hours in 2027-2029. The winners in each bucket look very different, so we split the list four ways:

• Integrators buy cells, wrap them in software and steel, and sell the finished system. Think "general contractor for batteries."
• Cell Makers manufacture the actual chemistry. Concentrated, Chinese, and enormous.
• LDES Pure-plays are the public companies betting on non-lithium chemistries. High risk, small caps, real revenue starting to show up.
• Private Bellwethers are the two private companies whose milestones move the whole LDES conversation.

An integrator doesn't invent the battery. It buys cells, engineers the enclosure, cooling, inverters, and controls, and then runs the software that decides when to charge and when to sell. It's a low-margin, high-volume business where the moat is trust: utilities want to know the box works in year twelve, not just year one.

Fluence Energy NASDAQ: FLNC

Fluence is the purest public bet on grid batteries, and right now it's a tale of two spreadsheets.

Spreadsheet one is demand. Q3 fiscal 2026 order intake hit a record $1.44 billion, nearly triple the year-ago quarter. Nine-month orders reached $2.7 billion, up 80%. Backlog is a company-record $6.4 billion, with about $2.2 billion expected to convert into fiscal 2027 revenue. The new engine is data centers: Fluence signed its first big behind-the-meter data center order (~$300 million) during the quarter, then picked up ~$550 million more from a hyperscaler portfolio in July. The data center pipeline alone is 16 GWh.

Spreadsheet two is execution. Q3 revenue came in around $650 million against expectations above $800 million, because new U.S. contract manufacturing lines ramped slower than planned. Adjusted gross margin compressed to 5.9% from 15.4%, the company swung to a $44 million net loss, and management cut full-year guidance. The stock sits more than 50% below its 52-week high.

Order books this size don't lie about demand. They lie about timing. If the factory problems are a two-quarter stumble, this is the cheapest way to own the U.S. storage buildout. Watch the FEOC-compliant cell supply agreement management says is coming, and watch the fiscal Q4 print for whether the ramp actually ramped.

Tesla NASDAQ: TSLA

You don't buy Tesla for the batteries. But the batteries are quietly the best business inside it.

Q2 2026 deployments were 13.5 GWh, up 40% year over year and 53% sequentially. Trailing-twelve-month deployments hit a record. Installed Megapack production capacity now stands at 40 GWh in Lathrop, California, plus 20 GWh in Shanghai, with a Texas Megafactory in commissioning and slated to start production this year. Megapack 3 and the Megablock product launched in 2025 and are moving into volume.

Storage is a slice of a company whose valuation swings on robotaxis, humanoid robots, and whatever the CEO tweeted that morning. Tesla itself warns that Megapack deployments are lumpy quarter to quarter and depend on cell supply. So treat this as a diversified way to own the #1 or #2 global integrator.

Sungrow SZSE: 300274, HKEX listing pending

Sungrow started life as a solar inverter company in Hefei. It turned that power-electronics know-how into grid storage, and in 2026 Wood Mackenzie ranked it the world's #1 BESS integrator, ahead of Tesla and CATL.

Storage passed inverters as Sungrow's largest business in 2025, with storage revenue up 49% to about CNY 37.3 billion on 43 GWh shipped. The 2026 target is 60-65 GWh. Cumulative shipments crossed 93 GWh by December. Its grid-forming PowerTitan 3.0 platform launched in Madrid in January, aimed squarely at European utilities that need inertia and black-start services, with a 10 GWh Europe delivery target for the year.

This is a Shenzhen-listed A-share, which most U.S. brokerages can't touch directly. Sungrow's first Hong Kong listing attempt lapsed; it refiled in April 2026. If the H-share lists, this becomes far easier to own. It also carries the flip side of the policy risk below: Chinese-origin systems face roughly 55% tariffs entering the U.S. and can't qualify for the federal tax credit, so Sungrow's growth is Europe, the Middle East, Australia, and everywhere else.

Energy Vault NYSE: NRGV

Energy Vault is the comeback story of this list. It went public on gravity storage (stacking giant bricks with cranes), got laughed at, and rebuilt itself as a lithium integrator that also owns and operates some of its own projects.

Q2 2026 revenue was $17.4 million, up 104%, driven by Australian projects. Adjusted gross margin expanded 900 basis points to 38.6%. Backlog jumped 40% quarter over quarter to roughly $2 billion, more than double a year ago, including a 1.25 GWh agreement with a hyperscaler. Management raised full-year revenue guidance to $270-310 million and GAAP gross margin guidance to 20-25%, and pointed to roughly $180 million of annual run-rate EBITDA from its owned-asset portfolio within 18-36 months.

Still loss-making (adjusted EBITDA was negative $17 million), still small, and revenue recognition is back-half-loaded into Q4, so quarterly prints will be jumpy. But a stock in the $3-4 range with a $2 billion backlog and expanding margins is the kind of setup this newsletter exists to flag.

The cell is the battery. Everything else is packaging. And cell manufacturing is one of the most concentrated industries on Earth: the top ten storage cell suppliers, all Chinese, shipped over 80% of global volume in the first half of 2026. There's one name you need here.

CATL SZSE: 300750 / HKEX: 3750

CATL is the largest battery maker in the world and has led global energy storage battery shipments for five straight years. In the first half of 2026 it shipped 125 GWh of ESS cells, up 81%, for a 27.1% global share, and storage is now nearly a fifth of company revenue. It raised roughly $5 billion in a Hong Kong share placement in April that sold out within an hour.

Two things make CATL more interesting than "big Chinese battery company." First, the cell format race: CATL moved a 587 Ah cell into production, and the industry expects 500 Ah+ cells to be the dominant utility-scale format by 2027. Bigger cells mean cheaper systems, which is bad for LDES pitches built on "lithium is expensive." Second, sodium-ion: CATL is partnering with Alfen to deploy 5 GWh of sodium-ion storage across Europe, a chemistry with no lithium and no thermal-runaway problem.

CATL cells are the textbook case of what U.S. FEOC rules exclude. That doesn't hurt CATL much (the rest of the world is buying), but it does mean the U.S. buildout is a market it mostly watches from outside.

These are the public companies that decided lithium can't do everything and built a different chemistry. The bull case is simple: when a grid needs 10+ hours of storage, stacking more lithium gets expensive fast, and these systems use cheap, non-flammable materials like zinc, iron, salt, and water. The bear case is also simple: they have to out-manufacture an industry that just doubled its output in twelve months.

Eos Energy NASDAQ: EOSE

Eos makes zinc-based batteries (its Z3 module) rated for 3 to 12 hours of discharge, built in Pennsylvania from materials that don't come from China. That last part matters a lot more than it did two years ago.

Q2 2026 revenue hit a record $68.8 million, up 351% year over year, and first-half revenue of $125.7 million already beat all of 2025. Backlog reached $807 million (3.4 GWh), and the commercial pipeline is $24.6 billion. Recent wins include a 750 MWh supply agreement in Germany, Austria, and Switzerland, a $100 million Blanquilla order, and a defense contract tied to the Golden Dome program. The company's Frontier Power USA joint venture, which develops projects that buy Eos batteries, raised $263 million to fund about $1 billion of deployments and delivered roughly 80% of Q2 revenue.

Gross margin was negative 71%, adjusted EBITDA was negative $71 million, and 2026 revenue guidance was narrowed to $300-350 million because Line 1 goes down while manufacturing consolidates. The fleet's round-trip efficiency is about 78%, well under lithium's 90%+. Cash was $364 million at quarter end. This is a company where the order book says "winner" and the income statement says "prove it." Q4 is when management says the cost benefits show up. Hold them to it.

ESS Tech NYSE: GWH

ESS builds iron-flow batteries out of iron, salt, and water in Oregon. The technology just got a public validation: a Burbank Water and Power evaluation, done for the American Public Power Association, concluded the system works as intended and has a role in a utility's storage mix. ESS also commissioned two systems for Turlock Irrigation District, acquired the assets of Germany's VoltStorage, and signed with Alsym Energy to bring non-flammable sodium-ion into its lineup.

Now the hard part. Q1 2026 revenue was $128,000. Liquidity was about $21.5 million after a $15 million registered direct offering at $1.75 a share. New CEO Drew Buckley describes the company as resetting around execution and capital discipline, which is what you say when the last plan didn't work. The NYSE has also put the company on notice for trading below $1.00 and delisted its warrants in June 2026, so a reverse split or a move off the big board is a live possibility.

It stays on the list because the tech is now third-party validated, the domestic-content angle is exactly what FEOC rules reward, and the market cap is small enough that a single large "Energy Base" contract changes the story. Just size it like the lottery ticket it is.

Publicly tradable LDES is thin, and the two most important LDES companies on the planet aren't listed. You can't buy them (unless you're accredited and shopping secondary markets), but their milestones set the tone for everything above.

Form Energy (Private)

Form Energy makes iron-air batteries that work by rusting and un-rusting iron. They're slow, heavy, and can discharge for 100 hours, which is the point. Nobody else is close on multi-day storage.

The 2026 headline was Google. In February, Google agreed to pay about $1 billion for a 300 MW / 30 GWh Form system, installed by Xcel Energy, to help power a data center in Pine Island, Minnesota alongside 1.4 GW of wind and 200 MW of solar. It's Form's first direct data center deployment and the largest battery by energy capacity ever announced. Deployment is expected to begin in 2028, and the plan still needs Minnesota Public Utilities Commission approval. In March, FuturEnergy Ireland signed for a 10 MW / 1,000 MWh system, Form's first international project, targeting 2029.

Form has raised roughly $1.4 billion to date, including a $405 million Series F led by T. Rowe Price with GE Vernova participating. Reporting in February said it's raising another $500 million and has IPO plans for 2027. Form Factory 1 in Weirton, West Virginia is targeting 500 MW of annual production capacity by 2028. If the IPO lands, it will be the first pure LDES listing at scale, and it will reprice Eos and ESS overnight, in one direction or the other.

Hydrostor (Private)

Hydrostor is the mechanical-storage bet. Its advanced compressed-air system pumps air into an underground cavern, uses the pressure to push water up to a surface reservoir, and reverses the process to spin a turbine when the grid needs power. Eight-plus hours, no exotic chemistry, and a design life measured in decades.

The flagship is Willow Rock in Kern County, California: 500 MW / 4,000 MWh, roughly $1.5 billion, with a $1.76 billion conditional DOE loan guarantee from January 2025 and a 25-year, 200 MW offtake deal with Central Coast Community Energy. It received its final power plant license in March 2026. On August 12, Hydrostor closed a $230 million equity round (Goldman Sachs Alternatives, CPP Investments, Canada Growth Fund, plus new money from Baker Hughes and Hatch) to carry Willow Rock to financial close in late 2026, with groundbreaking planned for later this year. A second project, 200 MW / 1,600 MWh in Broken Hill, Australia, is also near construction.

No IPO chatter yet. But a project this size hitting financial close would be the single biggest proof point for non-battery LDES in the U.S., and it's the catalyst we'd watch most closely between now and year-end.

Washington could rewrite the math. Storage kept its 30% investment tax credit under last year's budget law, with a phase-out that doesn't start until 2033. That's the good news. The bad news is the Foreign Entity of Concern rule attached to it. Starting with projects that began construction in 2026, more than 55% of manufactured-product cost must come from non-Chinese sources, and that threshold rises every year. Miss it and the project gets 0% instead of 30-40%, with recapture risk on top. Stack on the Section 301 tariff on Chinese batteries, which jumped from 7.5% to 25% on January 1, plus other layers that bring the effective rate on Chinese BESS to roughly 55%. About three-quarters of U.S. lithium battery imports come from China. Treasury's interim FEOC guidance still leaves real questions open. If final rules turn out stricter, or if credits get revisited, U.S. project economics wobble and every integrator on this list feels it. Watch for final FEOC guidance from Treasury, and whether Fluence and others lock down compliant cell supply at a sane price.

Or the factories don't ramp. Every company on this list has demand. What they don't have yet is execution. Fluence missed by $90 million because new lines came up slow. Eos runs a negative 71% gross margin. ESS has a quarter's worth of cash. Meanwhile lithium cells keep getting cheaper and bigger, which erodes the "lithium is too expensive for long duration" argument that LDES was built on. Watch gross margin, not revenue. A storage company that can't make money at $6 billion of backlog won't make it at $12 billion either.

The next 12 to 18 months are about conversion: turning record backlogs into shipped gigawatt-hours and turning LDES pilots into financed projects. Here's the calendar we're watching.

• Fluence fiscal Q4 (November 2026) and the fiscal 2027 guide. About $2.2 billion of backlog is earmarked for fiscal 2027. If the U.S. manufacturing ramp is fixed, this is the pivot. If it isn't, expect another guide cut.
• Eos Q4 2026 results. Management says field-cost pressure fades and consolidation savings arrive in Q4. Gross margin has to move meaningfully toward zero.
• Tesla Megafactory Texas start of production (targeted this year) plus Q3 and Q4 deployment prints. Watch for full-year 2026 deployments to beat 2025's total.
• Hydrostor Willow Rock financial close and groundbreaking (late 2026). The first shovel in the ground for a $1.5 billion compressed-air plant is a bigger LDES headline than any earnings release.
• Form Energy's $500 million round and 2027 IPO prep, plus Minnesota PUC approval of the Google/Xcel package. Any S-1 filing instantly makes Form the reference valuation for the whole LDES category.
• Sungrow's Hong Kong listing. If the H-share prices, U.S. investors get a clean way to own the world's #1 integrator.
• Energy Vault's Q4. Most of its 2026 revenue is scheduled to land in that one quarter, and the 1.25 GWh hyperscaler contract begins delivering. Hit the $270-310 million guide and this stock re-rates.
• Treasury's final FEOC rules. The single biggest swing factor for U.S. storage economics, and it isn't on any company's earnings calendar.

Storage was the "boring" part of the energy transition for a decade. Then Google spent a billion dollars on a battery that takes four days to charge.