2026 heat-pump efficiency by brand: which makers top the EPREL list?

How we filtered EPREL listings and grouped models by brand

This comparison uses the current EPREL snapshot exposed through the Househeating Pulse catalog and market index. The registry snapshot contains 60,989 heat-pump models from 777 manufacturers as of 2026-05-17 (market_index_snapshot / Househeating Pulse · Market Index v1, computed from EPREL Public API). Across that full set, the average SCOP is 4.55 (market_index_snapshot / Househeating Pulse · Market Index v1, computed from EPREL Public API).

For the brand ranking, the relevant universe is the top 20 manufacturers by EPREL listing count, taken from the leaderboard hub and the live EPREL catalog. Those 20 brands are ranked by model count, but each entry also includes an average SCOP, which makes it possible to compare scale against mean efficiency (brand_share / EPREL Public API · brand-share aggregation).

A methodological limitation matters here. The corpus does not include per-brand breakdowns for refrigerant share, heat-pump type mix, or price bands. It also does not provide brand-level lists of standout models for the top-20 brands. So this article can answer the average-SCOP and catalog-depth questions directly, and it can say what the overall EPREL market looks like by type and refrigerant, but the registry extract supplied here does not record enough brand-detail data to quantify, for example, what share of Daikin Europe N.V. listings use R290 units versus R32 units.

For readers who want the underlying mechanics, the relevant Househeating Pulse notes sit in the methodology reference, while the official database remains the EU’s EPREL platform.

The 2026 brand leaderboard: who leads on average SCOP?

Among the top 20 EPREL-listed heat-pump brands by model count, the highest average SCOP belongs to LG Electronics Deutschland GmbH at 4.93 across 405 models (brand_share / EPREL Public API · brand-share aggregation). The top five on average SCOP are:

All figures above are from the same top-20 brand ranking (brand_share / EPREL Public API · brand-share aggregation).

The gap between first and fifth is narrow: 4.93 minus 4.66 equals 0.27 SCOP points (brand_share / EPREL Public API · brand-share aggregation). That is a useful result in itself. The top of the table is not separated by a full efficiency class jump; it is clustered.

The ranking also looks different from a pure market-share league table. Daikin Europe N.V. is the largest manufacturer in EPREL with 14,668 listings and a 24.05% share, but its average SCOP is 4.44 rather than top-five territory (brand_share / EPREL Public API · brand-share aggregation). Mitsubishi Electric Europe B.V. is second by volume with 5,575 models and a 9.14% share, and posts a higher 4.51 average SCOP (brand_share / EPREL Public API · brand-share aggregation). Bosch Thermotechnik GmbH, by contrast, combines top-three efficiency with genuine scale: 3,602 models, 5.91% of listings, and a 4.69 average SCOP (brand_share / EPREL Public API · brand-share aggregation).

At the lower end of the top-20 efficiency range, Toshiba Carrier Europe S.A.S averages 3.93 across 497 models (brand_share / EPREL Public API · brand-share aggregation). That means the spread from the best-performing top-20 brand to the weakest is 1.00 SCOP point, from 4.93 to 3.93 (brand_share / EPREL Public API · brand-share aggregation).

Model depth versus efficiency: big catalogs or a few elite lines?

For buyers and journalists, the practical question is whether high average SCOP is coming from a broad portfolio or a compact specialist range.

The clearest broad-portfolio efficiency case in the top 20 is Bosch Thermotechnik GmbH: 3,602 models with an average SCOP of 4.69 (brand_share / EPREL Public API · brand-share aggregation). Ariston SpA is another large-range performer, with 2,618 models and a 4.66 average SCOP (brand_share / EPREL Public API · brand-share aggregation). Those are not niche listings.

At the opposite extreme, several efficiency leaders have much smaller catalogs. LG Electronics Deutschland GmbH reaches the highest average SCOP, 4.93, but across 405 models only (brand_share / EPREL Public API · brand-share aggregation). STIEBEL ELTRON GmbH & CO. KG follows at 4.84 across 433 models (brand_share / EPREL Public API · brand-share aggregation). GORENJE GOSPODINJSKI APARATI D.O.O. posts 4.67 across 440 models (brand_share / EPREL Public API · brand-share aggregation). These are still meaningful lineups, but they are not broad enough to describe as market-wide assortments on the scale of Daikin or Mitsubishi Electric.

That contrast becomes sharper when set against the volume leaders. Daikin’s 14,668 models are more than 36 times LG’s 405 (brand_share / EPREL Public API · brand-share aggregation). Mitsubishi Electric’s 5,575 models are around 1.55 times Bosch’s 3,602, yet Bosch still leads Mitsubishi on average SCOP by 0.18 points, from 4.69 versus 4.51 (brand_share / EPREL Public API · brand-share aggregation).

So the answer to the portfolio question is mixed. Some leaders are small-to-mid-sized and may be lifted by narrower ranges. But at least two brands in the top-five efficiency group — Bosch and Ariston — also have large catalogs. That is a stronger signal for mainstream buyers than a very high average derived from only a few hundred entries.

For context, the market itself is dominated by a few large vendors. The top three by count — Daikin, Mitsubishi Electric, and JOHNSON CONTROLS HITACHI AIR CONDITIONING EUROPE SAS, SUCURSAL EN ESPAÑA — account for 14,668, 5,575, and 5,207 models respectively (brand_share / EPREL Public API · brand-share aggregation). Hitachi’s average SCOP is 4.18, which places it well below the full-market average of 4.55 (brand_share / EPREL Public API · brand-share aggregation; market_index_snapshot / Househeating Pulse · Market Index v1, computed from EPREL Public API).

Refrigerants and product mix: who is betting on R290, and where?

This is where the supplied registry cut runs out of detail. The corpus does not provide per-brand refrigerant shares or per-brand type shares for the top 20 manufacturers. So the article cannot quantify which of the leading brands is most concentrated in R290 products, which still leans on R32 products, or whether a given brand’s average SCOP is driven mainly by air-water units, air-air units, ground-water units, water-water units, or heat-pump water heaters.

What the market-wide snapshot does show is the shape of the current EPREL population. Air-water is the largest category at 30,452 models, followed by air-air at 21,065 and hp-water-heater at 9,228; ground-water has 213 and water-water 31 in this snapshot (market_index_snapshot / Househeating Pulse · Market Index v1, computed from EPREL Public API). On refrigerants, R32 accounts for 13,935 listed models, while R290 accounts for 537 (market_index_snapshot / Househeating Pulse · Market Index v1, computed from EPREL Public API). The market index also reports a natural refrigerant share of 3.27% (market_index_snapshot / Househeating Pulse · Market Index v1, computed from EPREL Public API).

That means any editorial claim that the leading brands’ efficiency is specifically explained by R290 concentration would require a deeper manufacturer-level extract than the one available here. Readers can track those patterns in the live manufacturer directory, the market index snapshot, and the refrigerants reference, but the present corpus does not support brand-by-brand percentages.

The same applies to price bands. The registry excerpt supplied for this article does not record price-band positions for the leading brands, so it cannot support a numeric ranking of “premium-only” versus “mainstream-priced” lineups.

The standout models: which brands own the top of the EPREL table?

The top of the overall SCOP table is not owned by any of the biggest top-20 brands. The single highest-SCOP model in the supplied corpus is Risch Kälte- und Klimatechnik GmbH OH I 4esr TWW W/W at 7.0 SCOP (top_models / EPREL Public API via Househeating Pulse catalog). Next come Waterkotte GmbH EcoTouch DS 5034.5 T (water/water) and Hoval Aktiengesellschaft 42 -Thermalia® twin (26) GW, both at 6.97 (top_models / EPREL Public API via Househeating Pulse catalog). Master Therm tepelná čerpadla s.r.o. AQ30I-0WW also reaches 6.97, and Waterkotte GmbH CTC EcoTouch 525 (water/water) matches that level as well (top_models / EPREL Public API via Househeating Pulse catalog).

Within the top 20 individual models, Waterkotte appears 5 times, Master Therm 3 times, Hoval 2 times, Walter Bösch 2 times, and NIBE 3 times (top_models / EPREL Public API via Househeating Pulse catalog). The SCOP range inside that top-20 model list runs from 7.0 down to 6.8 (top_models / EPREL Public API via Househeating Pulse catalog). Most are classed APPP for heating (top_models / EPREL Public API via Househeating Pulse catalog).

This creates an important distinction between model-level excellence and brand-level average performance. A manufacturer can own several of the very best units in EPREL without appearing anywhere near the top-20 list by listing volume. Conversely, the large consumer-facing brands dominate the market by number of entries, not by occupying the first page of the absolute SCOP leaderboard. Readers can inspect the broader pattern in the top SCOP leaderboard and the air-to-water SCOP ranking.

The corpus does not provide the best-scoring model for each top-20 brand, so it is not possible here to calculate how much each major brand’s standout model exceeds its own brand average.

What buyers should infer: efficiency, breadth and mainstream pricing

The cleanest reading of the 2026 EPREL data is that buyers should separate three ideas that are often blurred together: average brand efficiency, catalog breadth, and absolute best-case model performance.

On average SCOP among the top-20 brands, the leaders are LG at 4.93, STIEBEL ELTRON at 4.84, Bosch at 4.69, Gorenje at 4.67, and Ariston at 4.66 (brand_share / EPREL Public API · brand-share aggregation). On sheer catalog depth, the leaders are Daikin with 14,668 models, Mitsubishi Electric with 5,575, Hitachi with 5,207, Bosch with 3,602, and Ariston with 2,618 (brand_share / EPREL Public API · brand-share aggregation).

The overlap matters most. Bosch and Ariston are the strongest examples in this corpus of brands that pair high average efficiency with broad portfolios rather than only a narrow line. Daikin remains the clear scale leader, but its 4.44 average SCOP sits below the full-market average of 4.55 by 0.11 points (brand_share / EPREL Public API · brand-share aggregation; market_index_snapshot / Househeating Pulse · Market Index v1, computed from EPREL Public API). Mitsubishi Electric at 4.51 is closer to the market mean but still not among the top efficiency leaders (brand_share / EPREL Public API · brand-share aggregation).

What cannot be said from this corpus is which of those brands also sits in mainstream EPREL price bands, because the supplied registry cut does not include price data. Nor can the article prove a brand-level R290 strategy from the available probes. That absence is worth stating plainly, especially because marketing language often implies more precision than the database actually supports.

For a buyer shortlisting real products rather than brand reputations, the sensible workflow is to start with the full heat-pump catalog, then filter by manufacturer, type, refrigerant and energy class, and finally test economics with the heat-pump payback calculator and capacity assumptions with the sizing calculator. Brand averages are useful. They are not a substitute for checking the exact unit.

Sources

• EPREL Public API · brand-share aggregation — snapshot as of 2026-05-17
• EPREL Public API via Househeating Pulse catalog — top-models snapshot as of 2026-05-17
• Househeating Pulse · Market Index v1, computed from EPREL Public API — snapshot as of 2026-05-17
• brand_detail probe failures for manufacturer slugs daikin, midea, and viessmann — no as-of date provided; data unavailable in supplied corpus

Continue reading

• Heat pump payback calculator — Compare upfront cost and running savings assumptions before narrowing a shortlist.
• Heat pump sizing guide — A practical route from heat loss and flow temperature to a sensible capacity range.
• R290 vs R32 heat pumps — What the refrigerant choice changes in efficiency, safety constraints and future compliance.
• How to read EPREL heat-pump labels — A quick decoder for SCOP, energy class, climate conditions and test points.