500,000 Battery Tests Rank Hyundai Ioniq 5 First for Capacity, Ahead of Tesla

The Hyundai Ioniq 5 ranked first for median retained battery capacity in a comparison built from more than 500,000 electric-vehicle battery tests. At 50,000 kilometers, or roughly 31,000 miles, tested Ioniq 5 batteries had a median state of health of 97.13%. Tesla’s Model Y ranked 14th at 94.42%, while the Model 3 placed 18th at 93.77%.

That puts Hyundai ahead in this particular 20-model ranking, but it does not establish that every Ioniq 5 battery will outperform every Tesla battery. AVILOO, the Austrian battery-diagnostics company that compiled the results, reported substantial variation among individual vehicles of the same model. The table’s model medians describe the middle of each tested population not the condition of a specific car.

Every model retained a median of more than 90%

The broader result is reassuring for EV owners. None of the 20 models recorded a median capacity decline greater than 10% over the first 50,000 kilometers. Battery state of health measures remaining capacity relative to the battery’s original capacity, so a 95% result indicates that the pack can store approximately 95% of what it could when new.

The Mercedes EQA ranked second with a 96.56% median, followed by the Hyundai Kona at 96.27%. At the other end, the Renault Zoe ranked 19th at 91.64%, and the Nissan Leaf ZE1 finished 20th at 91.23%. The gap between the first- and last-ranked medians was 5.9 percentage points.

AVILOO compiled tests conducted from 2022 through 2026 across Europe, North and South America, Asia and Australia. That geographic reach and the total number of tests give the comparison unusual scale, but several details needed for a controlled model-to-model assessment were not provided. These include the number of vehicles representing each model, their ages, battery variants, climates, charging histories and the statistical definition of the published lower and upper bounds.

Pack size is only part of the system

The comparison associates smaller packs with more frequent charging cycles and potentially greater capacity loss. For the same distance traveled, a smaller battery generally must process its usable energy more times than a larger one. The Zoe and Leaf were cited as examples of that broad pattern.

But pack size cannot by itself explain the ranking. The Model Y’s 14th-place result, despite its relatively large battery, illustrates the limits of a single-variable explanation. Battery chemistry, thermal control, software-defined operating buffers, charging power, climate and total energy throughput can all affect degradation. The dataset does not isolate those variables well enough to assign causes to the positions of individual models.

A separate Geotab analysis of more than 22,700 EVs reinforces why operating history matters. It found higher average degradation among vehicles using high-power direct-current fast charging more frequently, as well as a smaller climate-related difference between hot and mild operating environments. Those findings do not explain AVILOO’s model order, but they show why two nominally identical cars can reach the same mileage with different remaining capacity.

A ranking is not an inspection

For an owner, capacity loss can translate into less usable range. It can also affect confidence in a used vehicle’s value and fitness for a particular commute. Yet the practical effect depends on the car’s original usable capacity, efficiency, operating temperature and driving conditions; a state-of-health percentage is not a direct range guarantee.

Warranty thresholds also answer a different question. Hyundai’s U.S. warranty information lists 10-year or 100,000-mile coverage for hybrid and EV batteries and system components, with capacity language tied to 70% of original capacity. Tesla’s current warranty page lists eight-year battery and drive-unit coverage for Model 3 and Model Y variants, with mileage limits varying by version and a minimum 70% capacity-retention provision. A battery can therefore remain well above a warranty threshold while still having measurable capacity loss relevant to daily use.

The 500,000-test comparison supports two conclusions at once: early median degradation was modest across all 20 models, and the Ioniq 5 led the model-level ranking. But because the reported bounds show meaningful within-model variation and key sampling details remain unspecified the condition of an individual used EV still requires an individual battery assessment.

By Thomas Caldwell — AMI’s senior editor for mechanical and mobility engineering, covering vehicle electronics, systems integration, electrification, chassis systems, propulsion, and safety policy.