If you have been following the Donut Lab story, the last few weeks have finally made it easier to separate the serious part from the sensational part.

The company now has a more credible pack-level charging result inside a Verge motorcycle, a damaged-cell safety result that is at least directionally useful, and a new round of skeptical mainstream coverage that is asking the right questions. Taken together, that is enough to move the story forward. It is not enough to prove the entire Donut Lab case.

If you want the earlier context first, I wrote the original overview here: Donut Lab Solid-State Battery: 400 Wh/kg, Fast Charging, and the Evidence So Far. I also wrote a narrower follow-up focused on the third VTT-backed public test here: Donut Lab Battery Update: Verge Pack Charges to 80% in 12 Minutes.

What is actually new

The important part is that this is no longer just one company press release repeated three different ways.

Over the last few weeks, Donut Lab’s battery story has picked up three distinct new data points:

  • a pack-level charging test in a Verge TS Pro motorcycle
  • a damaged-cell safety and cycling result
  • a more skeptical mainstream read of the full claim stack

That combination matters because each one answers a different question.

The charging test asks whether the company can move beyond cell-level lab talk. The damaged-cell result asks whether the safety story still holds when things go wrong. The skeptical coverage matters because it forces the conversation back toward the unresolved claims that still define whether this is a genuinely disruptive battery or just an unusually well-packaged battery story. (3, 4, 5, 6, 7)

The Verge motorcycle result is still the most meaningful update

This is still the part that matters most.

Electrek and electrive both covered Donut Lab’s public pack-level charging test inside a Verge TS Pro. The standard 18 kWh pack reportedly sustained more than 100 kW of charging power for five minutes, moved from 10% to 50% in five minutes, reached 70% in just over nine minutes, and hit 80% in 12 minutes. Electrive also noted that the test used a public Alpitronic Hypercharger at Circle K and that the motorcycle was a previous-generation TS Pro fitted with the new model-year battery. (3, 4)

That is not trivial.

It is one thing to charge a single cell quickly in a controlled test setup. It is a more serious step to show a pack-level result in an actual vehicle, with real packaging constraints and a very limited thermal-management envelope. For a motorcycle, an air-cooled pack sustaining that kind of power is genuinely interesting. (3, 4)

At the same time, it is worth staying disciplined about what this does not prove. A pack charging result tells us that Donut Lab can show real high-power behavior outside the lab-bench-cell context. It does not prove the full production story, the long-term durability story, or the headline energy-density story.

The safety result is useful, but it is not a clean victory lap

The damaged-cell safety test is the other important recent development.

Electrive reported that Donut Lab took the same DL2 cell that had lost its vacuum during the earlier 100 C test and continued cycling it afterward. According to the reporting, the cell completed five standard 1C cycles and then more than 50 cycles at 5C, all without the kind of thermal runaway or fire event you would worry about in a conventional lithium-ion pouch cell with severe structural damage. (5)

That is a meaningful data point, especially because the whole Donut Lab pitch depends not just on charge speed and density claims, but also on the argument that the chemistry is materially safer.

But this is also where the more skeptical coverage becomes useful. Electrek’s March 23 follow-up makes an important point: the damaged cell is also the closest thing outside observers have seen to a real cycle-life-style result, and it is not flattering if you overread it. The cell’s capacity dropped sharply during the 5C cycling sequence, which is not surprising for a compromised sample, but it also means this test is not evidence for the much bigger durability claims. It is better read as a safety-and-failure-mode result than as a proof of endurance. (6)

That distinction matters. Donut Lab can reasonably point to the result as evidence that the cell did not fail catastrophically. It cannot reasonably point to it as confirmation of the broader cycle-life story.

The skeptical coverage is part of the story, too

The newer skeptical reporting is not just noise around the edges. It is one of the more useful parts of the current moment.

The Verge’s April 11 piece is valuable because it pulls the camera back. It treats the VTT-backed results as real progress while still emphasizing that the boldest claims remain unresolved: 400 Wh/kg, 100,000 cycles, and a commercial-scale manufacturing story that can survive contact with independent validation. It also highlights the more awkward parts of the testing sequence, including the damaged-cell incident and the unanswered chemistry questions. (7)

That is exactly the right framing.

At this point, I do not think the serious question is whether Donut Lab has shown anything. It clearly has. The more important question is whether the validated results are the early edge of a truly exceptional battery, or just the most believable subset of a much larger claim stack. The skepticism is healthy because it keeps those categories from getting blurred together.

What still is not proven

This part has not changed enough yet.

The claims that made Donut Lab famous in the first place are still the ones that need the cleanest proof:

  • roughly 400 Wh/kg energy density
  • around 100,000 cycles
  • scalable production
  • cost and manufacturability advantages over conventional lithium-ion

Those claims still matter because they are the difference between “this company has a very interesting fast-charging battery story” and “this company has built something that meaningfully changes the battery landscape.” Donut Lab is still publicly making those larger claims on its battery page, but the recent third-party-backed results still do not settle them. The Verge also reported that CEO Marko Lehtimaki described the widely repeated 100,000-cycle figure as a design target, not an experimentally validated result. That is exactly the kind of clarification this story still needs. (7, 8)

My take now

My view is a little firmer than it was a month ago.

I do not think Donut Lab has proved the whole case.

I do think it is getting harder to dismiss the company outright.

The reason is not that one miracle stat suddenly got confirmed. It is that the story keeps surviving the kinds of tests and scrutiny that would have been easy to fail if this were just launch-stage theater. The pack-level Verge result matters. The damaged-cell safety result matters. The skeptical coverage matters. Together, they make the current Donut Lab story stronger, but they also make the remaining evidence gap easier to see.

That is why I think the most honest way to read the latest developments is this:

Donut Lab looks more serious than it did at CES, but the biggest claims are still waiting for the kind of proof that would end the argument instead of extending it.