GM LMR Battery Platform Cuts Pack Size 20% for Mass-Market EVs

What GM announced

On October 21, General Motors used its GM Forward event in New York to show its fourth-generation electric vehicle platform. The platform is built on lithium manganese-rich (LMR) prismatic battery cells, which GM calls the world's first production LMR battery.24 The company said the architecture will go into its highest-volume EV segments. A spokesperson later told InsideEVs that this means crossover SUVs.21 Ultium Cells, GM's joint venture with LG Energy Solution, will build the cells in Spring Hill, Tennessee. Output is due to start in 2028.22

GM's main numbers describe the whole vehicle, not just the battery. The company says the platform makes vehicles 15% lighter, 20% more efficient and 25% better aerodynamically, while needing 20% less battery capacity overall.21 Its new drive units contain 40% fewer parts. A centralized electrical architecture uses 30% less wiring and offers 40 times more processing power.24 GM says the battery, motors and aerodynamics together should deliver about 20% more range per kilowatt-hour than its current vehicles.24

The event did not include a launch date or any model names. President Mark Reuss said only that the vehicles would arrive "very shortly."21 Interesting Engineering pointed out that the efficiency figures are still company targets and have not been independently verified. That caution applies to the whole announcement.

The chemistry bet

The pitch for LMR is simple. GM says the cells store about 33% more energy than lithium iron phosphate (LFP) cells at a similar cost, which allows a smaller, lighter pack.24 Manganese is the main ingredient. Compared with nickel-manganese-cobalt (NMC) cells, LMR uses less nickel and almost no cobalt.22 When GM first described the chemistry in 2025, MotorTrend reported a cathode mix of roughly 60–70% manganese, up to 2% cobalt, and nickel for the rest. Typical high-nickel NMC cells are about 85% nickel.7

The main technical problem with LMR has been voltage fade, a gradual drop in performance over repeated charging.22 GM battery chief Kurt Kelty said in February that the company had solved it.22 Earlier coverage said LMR cells have historically had much shorter lifespans.9 InsideEVs noted that GM still expects durability comparable to its NMC packs, including in hot and cold weather.8 Not everyone is satisfied. CleanTechnica noted that cold-weather performance and charging speed were not discussed in New York.23 GM did make those promises in 2025, saying LMR would charge faster than the NMC packs in its current trucks and perform better in the cold.7 For now, those claims rest on GM's word.

The most notable new detail is that GM now plans to use LMR alone on this platform. Reuss said Gen 4 will "start off exclusively with LMR."26 Chief product officer Sterling Anderson said engineers had first expected to need a second, longer-range chemistry, but improvements in LMR made that unnecessary.26

This is a real shift from GM's earlier messaging. In 2025, the company described LMR as a middle option between cheap LFP and high-nickel NMC, starting in full-size trucks and SUVs like the Silverado and Escalade IQ.13 Pointing it at crossovers instead suggests GM thinks the chemistry is ready for mainstream vehicles, not only a fix for heavy pickups that need large batteries.

Where the coverage diverges

Outlets interpreted the strategy very differently. CleanTechnica called it a big gamble. It argued that most of the industry is moving toward LFP or sodium-ion, and that GM's choice will either pay off or push the company further toward serious trouble.23 InsideEVs was more neutral. It treated LMR as a cost lever and noted that Ford's Universal EV platform, with a $30,000 pickup due late next year, appears to be further along.21

CleanTechnica's claim that no one else is using LMR also goes too far. Ford plans its own LMR-powered EVs by the end of 2029.22 Based on what has been announced, GM looks more like an early mover on the chemistry than a lone holdout.

The outlets also disagree slightly on what reaches the 2028 Cadillac Escalade IQ first. InsideEVs says that vehicle introduces the new compute and electrical architecture and GM's eyes-off driving system.21 One summary merged this with the platform debut.5 The more careful reading is that the Escalade IQ gets the new electronics in 2028, while GM has not said which nameplates will use the fourth-generation EV platform.21

The financial backdrop is hard to ignore. GM took $6 billion in EV-related charges in the fourth quarter of 2025 and converted its Orion, Michigan, plant to combustion vehicles.22 Another report puts its total EV retreat at $11.2 billion.25 Reuss's statement that GM is "not walking away" from EVs came with a qualifier: the company will be ready "when the U.S. market is ready."28 Our view is that the announcement is a cost-structure plan meant to keep GM's EV options open, not a sign of renewed growth in U.S. EV sales.

The grid-storage connection

The energy-storage side helps explain the timing. The same Spring Hill plant that will make LMR cells has already shifted part of its capacity to stationary storage. In March, Ultium Cells announced a $70 million retooling to make LFP cells for LG Energy Solution Vertech's energy storage systems, aimed at grid projects and AI data centers.11 The plant moved those lines from EV to storage production in less than five months, and furloughed workers came back.16 LG's North American president said the storage business had helped make up for weaker-than-expected EV demand.11

This matters for LMR because it shows the plant can switch between chemistries. Ultium says its manufacturing platform supports several chemistries and can follow demand.12 Spring Hill now makes LFP cells for storage and is preparing prismatic LMR cells for cars. That flexibility reduces GM's risk. If EV demand stays weak, the factory has other work. If demand recovers, it can build automotive cells.

GM has also described a clear division between chemistries. In June, Kelty said utilities and hyperscalers care about reliable, affordable power over long periods, not range or weight.14 That is why GM is developing sodium-ion cells for grid storage with startup Peak Energy, with trial production planned for 2028.20 Kelty compared sodium-ion to LMR, calling both early-stage chemistries with more room to improve than LFP, which he said is starting to plateau.14 So GM is betting on less-mature chemistries in both markets. LMR is its answer for vehicles, sodium-ion is its answer for the grid, and LFP covers storage in the meantime. One analysis described the broader industry as splitting into high-density chemistries for vehicles and low-cost or second-life chemistries for stationary storage, with GM among the first Western automakers pursuing both.15

The details of the LFP supply chain vary between reports. Ultium and Korean coverage say Spring Hill produces LG's cells for Vertech.1117 Some outlets describe GM supplying LFP cells from its Warren development center instead.1520 Reports of the start date also range from June to early July 2026.2216

The missing vehicle-to-grid link

One gap stands out. In June, GM turned on bidirectional charging through a firmware update and said the feature would become standard across its lineup.22 It also announced a PG&E program for grid balancing by 2030. EV Infrastructure News puts the goal at up to 52,000 vehicles, while another report says 130,000.2219 Yet the fourth-generation platform announcement does not mention vehicle-to-grid at all.22

That may just be a messaging decision. Still, a platform designed around "smarter energy management" and much more computing power would be an obvious place to make bidirectional charging a headline feature.24 A smaller, cheaper battery also has less spare capacity to offer the grid, though GM has not addressed that tradeoff.

The bottom line

GM is betting that a single in-house chemistry, used across an efficient vehicle, can do what other automakers are trying to do with LFP and very low build costs. The company says it tested more than 300,000 battery designs in simulation and went from first pack concept to program approval in one month. It also says its new development process will cut the usual five-year timeline in half.24 If LMR delivers the cost, cold-weather performance and durability GM claims, a battery 20% smaller than today's is a major advantage. If it does not, the Spring Hill plant can keep making storage cells. Either way, GM has limited the downside, and that may matter as much as the chemistry itself.