Your Brand has an AI reputation and it's shaping purchase decisions right now.

Swipe up to continue

Category · EV vehicles

Which EV brand would you recommend?

6 AI models · 5 answers per model · ranked by score

ordered by brand score: 100 pts for rank 1 → 20 for rank 5 · no mention = 0

  1. #1Tesla
    89
    brand score
    78first choice
    12345
  2. #2Hyundai
    75
    brand score
    61first choice
  3. #3Kia
    48
    brand score
    28first choice
  4. #4Rivian
    25
    brand score
    16first choice
  5. #5BMW
    22
    brand score
    14first choice
Show 6 more
  1. #6Ford
    19
    brand score
    14first choice
  2. #7BYD
    15
    brand score
    10first choice
  3. #8Nissan
    3
    brand score
    3first choice
  4. #9Chevrolet
    2
    brand score
    2first choice
  5. #10Volvo
    1
    brand score
    1first choice
  6. #11Lucid
    1
    brand score
    1first choice

Each bar covers the middle half of one model’s answers (Q1–Q3); the line inside it is that model’s median rank for the times it recommended the brand. The badge under a brand says how far apart the models are on it, and “first choice” scores rank 1 far above the rest — a brand can place well overall and still rarely lead an answer.

Swipe left for the full report

Category · EV vehicles

Model by model

Median rank each model gave each brand.

Scroll the table sideways to see every model →

claude-opus-5deepseek-v4.1-flashgemini-3.7-flashgpt-6-astrakimi-k3mistral-medium-3.5
Tesla2.01.01.03.01.01.0
Hyundai1.02.02.01.02.04.0
Kia3.03.03.02.03.0
Rivian5.05.04.04.03.0
BMW4.04.04.04.0
Ford5.04.05.05.04.54.0
BYD5.05.02.0
Nissan5.0
Chevrolet5.05.0
Volvo5.0
Lucid5.0
1st
2
3
4
5th

A dash means that model never named the brand in any of its runs. Deeper blue is better: rank 1 is the brand a model would recommend first.

Category · EV vehicles

Where the models disagree

5 of 11 brands split the panel. Each mark is one model's score for the brand, on the same 0–100 scale as the ranking.

  1. #2Hyundai

    mistral-medium-3.5 8 gpt-6-astra 100

  2. #7BYD

    claude-opus-5 0 mistral-medium-3.5 80 · 3 of 6 never named it

  3. #3Kia

    mistral-medium-3.5 0 gpt-6-astra 80 · 1 of 6 never named it

  4. #4Rivian

    gpt-6-astra 0 mistral-medium-3.5 60 · 1 of 6 never named it

  5. #5BMW

    deepseek-v4.1-flash 0 gemini-3.7-flash 44 · 2 of 6 never named it

  6. #1Tesla

    gpt-6-astra 60 mistral-medium-3.5 100

A hollow mark is a model that never named the brand in any of its runs, which scores 0. Agreement is not endorsement — a brand every model ignores equally agrees just as tightly as one they all rank first.

Category · EV vehicles

Named, and named first

Across → how frequently the panel names the brand at all.
Up ↑ how often the answers that name it put it first.

the default answera narrow favouritelisted, rarely ledthe long tail12345678
  1. 1Tesla100/63
  2. 2Hyundai87/37
  3. 3Kia73/0
  4. 4Rivian63/0
  5. 5BMW53/0
  6. 6Ford63/0
  7. 7BYD23/0
  8. 8Nissan17/0

The horizontal line sits at 20% — the rate a named brand would lead at if the models were picking one of its 5 slots at random. Above it they are choosing it first on purpose. Both figures average across models, so a thinly sampled model counts the same as a heavily sampled one.

Category · EV vehicles

What the models actually said

Models
6
Answers
30

Tesla leads with a score of 89 of 100, ranked by 6 of 6 models.

Tesla anchors the category, ranked by every model and scoring 88.67, but the panel divides on how much weight to give its well-documented caveats rather than on the caveats themselves. Four models put it first almost always; claude-opus-5 and gpt-6-astra pull it down, with gpt-6-astra sitting furthest away at a median rank 3, placing it behind Hyundai and Kia on interface and build-quality grounds.

Tesla

per-model scores 60100 of 100 · mean 89 across 6 models

The sharpest genuine split in the field is Hyundai, whose per-model scores run the full width from 8 to 100. Claude-opus-5 and gpt-6-astra rank it first outright, three models settle it second, and mistral-medium-3.5 names it just once at rank 4. The high scorers converge on the E-GMP platform and value story, so the divide is one of enthusiasm rather than of substance.

gpt-6-astra
100
rates it highest
mistral-medium-3.5
8
rates it lowest

Hyundai · 92 points apart on a 0–100 scale

Kia tracks Hyundai closely as its E-GMP near-peer, but the divergence there is coverage, not sentiment — gemini-3.7-flash names it in only 40% of runs while others hold it steadily at rank 3, and no model ranks it first. Rivian and BMW sit in similar territory: both are read consistently on the merits, with the spread driven by how far each model lets caveats (young-company risk for Rivian, price for BMW) pull the brand down, and by silent models scoring zero.

The most instructive disagreement is BYD, where mistral-medium-3.5's rank-2 conviction collides with two models placing it fifth. All three agree on the battery and value strengths; they split on whether limited Western availability should cap the ranking — a geography question, not an engineering one.

Across the field the models rarely disagree about what a brand is; they disagree about how heavily to price its known weaknesses.

The lower tier — Ford, Nissan, Chevrolet, Volvo, Lucid — shows "agreement" that is mostly omission. Ford is the exception, named by all six but never first, its rank-4-to-5 clustering a true consensus. The others register with one or two models each, so their low standard deviations reflect near-universal absence rather than a shared verdict.

The recalled source base is uniform enough that it explains little of the divergence: Car and Driverwww.caranddriver.com×32 and Edmundswww.edmunds.com×29 dominate nearly every brand, with Consumer Reportswww.consumerreports.org×23 recurring on reliability points. Because the same outlets sit behind both high and low placements, any link between a specific source and a model's harsher or softer read is a hypothesis the data does not confirm — the models appear to weight shared evidence differently rather than to draw on different evidence.

Category · EV vehicles

What shaped the answers

37 sources across 431 references, grouped by site from 141 recalled names. The top 5 carry 66% of them.

Car and Driver×92 · 21%

caranddriver.com

Sources are what each model recalled as having shaped its view — not verified citations. A model without web access reconstructs a reference from memory, so a link may not lead where the model thought it did.

Category · EV vehicles
#1

Tesla

89
brand score
78first choice
12345

Summary

Tesla scores 88.67 of 100 on Borda and was ranked by all 6 of the 6 models, placing it among the strongest recommendations in the category. The panel reaches broad agreement on its merits: the recurring theme across every model is the Supercharger network as the anchor of an EV ownership ecosystem, paired with energy efficiency, range, and over-the-air software updates. Where models diverge, they do so on how much weight to give the widely noted caveats — inconsistent build quality, minimalist touchscreen-heavy controls, and uneven service — rather than on whether those caveats exist.

Tesla

per-model scores 60100 of 100 · mean 89 across 6 models

That split is visible in the per-model placements. Four models (deepseek-v4.1-flash, gemini-3.7-flash, kimi-k3, mistral-medium-3.5) rank it first most or all of the time, treating build quality as a minor caveat within a seamless ecosystem. Claude-opus-5 settles it at a median rank 2, framing it as the top pick specifically for charging convenience and long-distance travel, while gpt-6-astra sits furthest away at a median rank 3, consistently ranking it below Hyundai and Kia because the interface and fit-and-finish trade-offs make it, in its words, a less universal recommendation. The recalled sources cluster consistently under these themes — Consumer Reportswww.consumerreports.org×23 and Edmunds recur behind the reliability and ownership caveats, while EPA fueleconomy.gov and InsideEVs underpin the efficiency and range claims.

Tesla remains the benchmark for EVs due to its Supercharger network, long range, and advanced software.

deepseek-v4.1-flash

I rank it below Hyundai and Kia because its touchscreen-heavy controls, reported build-quality inconsistencies, and uneven service experiences are meaningful trade-offs.

gpt-6-astra

Per model summary

  • deepseek-v4.1-flash
    brand score
    96
    first choice
    90
    median rank
    1.0
    present in
    5/5

    Repeatedly positions Tesla as the EV benchmark for most buyers based on its Supercharger network, range, performance, and over-the-air updates, while noting mixed build quality and service or reliability concerns.

  • gemini-3.7-flash
    brand score
    100
    first choice
    100
    median rank
    1.0
    present in
    5/5

    Consistently calls Tesla the industry benchmark for its Supercharger network, energy efficiency, software integration, and OTA updates, treating variable build quality as a minor caveat within a seamless ownership ecosystem.

  • kimi-k3
    brand score
    100
    first choice
    100
    median rank
    1.0
    present in
    5/5

    Repeatedly describes Tesla as the most complete and easiest EV to own thanks to efficiency, Supercharger access, software/OTA updates, competitive pricing, and resale value, while citing inconsistent build quality and service.

  • mistral-medium-3.5
    brand score
    100
    first choice
    100
    median rank
    1.0
    present in
    5/5

    Consistently frames Tesla as the leader in EV technology, battery innovation, and charging infrastructure via the Supercharger network, emphasizing performance, range, and software with no drawbacks noted.

  • claude-opus-5
    brand score
    76
    first choice
    47
    median rank
    2.0
    present in
    5/5

    Consistently frames Tesla as the top pick for charging convenience and long-distance travel due to its mature Supercharger network, efficiency, range, and software/OTA updates, while flagging inconsistent build quality, minimalist interiors, and variable service.

  • gpt-6-astra
    brand score
    60
    first choice
    33
    median rank
    3.0
    present in
    5/5

    Highlights Tesla's efficiency, integrated route planning, and convenient charging for road trips, but consistently ranks it below Hyundai and Kia due to touchscreen-heavy controls, fit-and-finish variability, and uneven service.

Sources per model

Each model’s own references for this brand, grouped by site. Tile area is that site’s share of the model’s references; tap one for the pages behind it.

deepseek-v4.1-flash6 sites · 15 references
Car and Driver×4 · 27%

caranddriver.com

  • ×2/
  • ×1/tesla
  • ×1named without a URL
gemini-3.7-flash4 sites · 15 references
Car and Driver×5 · 33%

caranddriver.com

kimi-k36 sites · 15 references
Consumer Reports×5 · 33%

consumerreports.org

mistral-medium-3.53 sites · 15 references
Consumer Reports×5 · 33%

no URL recalled

  • ×5named without a URL
claude-opus-56 sites · 15 references
Consumer Reports×5 · 33%

consumerreports.org

gpt-6-astra5 sites · 14 references
Car and Driver×5 · 36%

caranddriver.com

Sources are what each model recalled as having shaped its view — not verified citations. A model without web access reconstructs a reference from memory, so a link may not lead where the model thought it did.

Category · EV vehicles
#2

Hyundai

75
brand score
61first choice
12345

Summary

Hyundai lands at a Borda score of

Stat unavailable

and was ranked by all 6 models, yet the panel is far from unanimous. Per-model borda spans 8 to 100 with a standard deviation of 31.28, which the report labels "sharply split." Two models (claude-opus-5 and gpt-6-astra) placed it first in all five of their runs, three others (deepseek-v4.1-flash, gemini-3.7-flash, kimi-k3) settled it at a median rank of 2, while mistral-medium-3.5 named it only once at rank 4 for a coverage of 20% and a borda of 8. The recurring theme across the high scorers is Hyundai's E-GMP platform: 800V ultra-fast charging, competitive pricing, long range, and a generous warranty, framed as the best all-around value alternative to Tesla.

gpt-6-astra
100
rates it highest
mistral-medium-3.5
8
rates it lowest

Hyundai · 92 points apart on a 0–100 scale

The sources under these themes cluster on mainstream automotive review outlets — Car and Driver, Edmunds, MotorTrend, IIHS, and Consumer Reports recur across models — with award citations such as World Car Awards backing the "well-rounded" framing. The caveats that hold Hyundai just behind the top are consistent too: less mature software, historically weaker charging network access (partly offset by NACS adoption), and ICCU recalls that several models flag as worth verifying before purchase. Mistral's lone, lower placement rests on a similar read but positions Hyundai as reliable rather than cutting-edge.

It's the best all-around balance of value, tech, and dependability for most buyers.

claude-opus-5

While not as cutting-edge as Tesla or BYD, it provides reliable and well-rounded options.

mistral-medium-3.5

Per model summary

  • claude-opus-5
    brand score
    100
    first choice
    100
    median rank
    1.0
    present in
    5/5

    Consistently highlights the E-GMP platform's 800V ultra-fast charging, strong efficiency/range, competitive pricing, and excellent warranty, framing Ioniq 5/6 as the best all-around value, with NACS access and awards noted; ICCU recalls and software glitches mentioned as caveats.

  • gpt-6-astra
    brand score
    100
    first choice
    100
    median rank
    1.0
    present in
    5/5

    Frames Hyundai as the best all-around recommendation for balancing efficiency, comfort, value, and fast charging, while repeatedly cautioning that charging performance varies by model and buyers should verify recall completion.

  • deepseek-v4.1-flash
    brand score
    84
    first choice
    60
    median rank
    2.0
    present in
    5/5

    Emphasizes the E-GMP platform's 800V ultra-fast charging, long range, competitive pricing, and generous warranty, positioning Hyundai as a well-rounded value choice; a smaller charging network is noted as the main improving downside.

  • gemini-3.7-flash
    brand score
    80
    first choice
    50
    median rank
    2.0
    present in
    5/5

    Repeatedly centers on the advanced 800V E-GMP platform enabling ultra-fast charging at mainstream prices, paired with distinctive styling, generous warranty, and strong overall value.

  • kimi-k3
    brand score
    80
    first choice
    50
    median rank
    2.0
    present in
    5/5

    Focuses on the E-GMP 800V platform's very fast charging, distinctive design, long battery warranty, and price advantage, ranking Hyundai just behind Tesla due to less mature software and charging network access.

  • mistral-medium-3.5
    brand score
    8
    first choice
    5
    median rank
    4.0
    present in
    1/5

    Describes Hyundai as offering a diverse EV lineup with solid range and competitive pricing, reliable and well-rounded but not as cutting-edge as Tesla or BYD.

Sources per model

Each model’s own references for this brand, grouped by site. Tile area is that site’s share of the model’s references; tap one for the pages behind it.

claude-opus-56 sites · 15 references
Car and Driver×5 · 33%

caranddriver.com

gpt-6-astra4 sites · 11 references
Car and Driver×5 · 45%

caranddriver.com

deepseek-v4.1-flash5 sites · 15 references
Car and Driver×5 · 33%

caranddriver.com

gemini-3.7-flash5 sites · 15 references
MotorTrend×5 · 33%

motortrend.com

kimi-k35 sites · 15 references
Car and Driver×4 · 27%

caranddriver.com

mistral-medium-3.53 sites · 3 references
Hyundai Official Website×1 · 33%

hyundai.com

Sources are what each model recalled as having shaped its view — not verified citations. A model without web access reconstructs a reference from memory, so a link may not lead where the model thought it did.

Category · EV vehicles
#3

Kia

48
brand score
28first choice
12345

Summary

Kia lands at

Stat unavailable

on Borda, ranked by 5 of the 6 models, but the panel does not settle on a single reading. Per-model borda spans 0–80 with a standard deviation of 27.23, which the report labels "models split." The divide is largely one of coverage rather than sentiment: gpt-6-astra places it consistently at median rank 2 for a borda of 80, while gemini-3.7-flash ranks it in only 2 of 5 runs, holding coverage to 40% and borda to 24. The three remaining models cluster tightly at median rank 3, scoring between 60 and 64.

Kia

per-model scores 080 of 100 · mean 48 across 6 models

The recurring theme is that Kia is a near-peer to Hyundai on the shared 800V E-GMP platform, valued for the three-row EV9, fast charging, and warranty coverage, but held just behind on pricing at higher trims, inconsistent dealer experience, and availability. Those points recur across the models' recalled sources, with Car and Driver×10 and Edmunds×6 appearing most often, alongside award citations and reliability references such as J.D. Power, IIHS, and Consumer Reports. First-choice score sits at 27.78, and no model ranked Kia first in any run, consistent with the repeated framing of it as a strong second or co-equal rather than the lead pick.

Kia is a close second, offering compelling electric family vehicles with practical interiors, strong warranties, and excellent charging speeds on its newer dedicated EV platforms.

gpt-6-astra

Essentially a co-equal to Hyundai, ranked just behind on value per dollar.

claude-opus-5

Per model summary

  • gpt-6-astra
    brand score
    80
    first choice
    50
    median rank
    2.0
    present in
    5/5

    Kia is consistently a close second to Hyundai, strongest for families needing space and three-row seating, but placed behind because larger, better-equipped models become expensive and value depends on needs.

  • claude-opus-5
    brand score
    64
    first choice
    37
    median rank
    3.0
    present in
    5/5

    Kia is framed as a co-equal to Hyundai on the shared 800V E-GMP platform, praised for the three-row EV9, fast charging, and strong warranty, but ranked just behind Hyundai due to rising prices on top trims and inconsistent dealer experience.

  • deepseek-v4.1-flash
    brand score
    60
    first choice
    33
    median rank
    3.0
    present in
    5/5

    Kia is positioned as sharing Hyundai's E-GMP platform while adding bolder styling and value, with the three-row EV9 as a standout, though dealer markups and availability are noted drawbacks.

  • gemini-3.7-flash
    brand score
    24
    first choice
    13
    median rank
    3.0
    present in
    2/5

    Kia is highlighted for exceptional value, fast charging, and family-friendly practicality, with models consistently earning awards for ergonomics, range, and price-to-performance.

  • kimi-k3
    brand score
    60
    first choice
    33
    median rank
    3.0
    present in
    5/5

    Kia shares Hyundai's excellent E-GMP platform with strong styling and the standout three-row EV9, but ranks just below Hyundai due to higher pricing on top trims and inconsistent dealer/availability.

Sources per model

Each model’s own references for this brand, grouped by site. Tile area is that site’s share of the model’s references; tap one for the pages behind it.

gpt-6-astra2 sites · 10 references
Car and Driver×5 · 50%

caranddriver.com

claude-opus-57 sites · 15 references
Edmunds×4 · 27%

edmunds.com

deepseek-v4.1-flash8 sites · 15 references
Car and Driver×5 · 33%

caranddriver.com

  • ×3/
  • ×1/kia
  • ×1named without a URL
gemini-3.7-flash4 sites · 6 references
Edmunds×2 · 33%

edmunds.com

kimi-k37 sites · 15 references
Car and Driver×3 · 20%

caranddriver.com

Sources are what each model recalled as having shaped its view — not verified citations. A model without web access reconstructs a reference from memory, so a link may not lead where the model thought it did.

Category · EV vehicles
#4

Rivian

25
brand score
16first choice
12345

Summary

Rivian lands at a modest

Stat unavailable

and is named by 5 of the 6 models, yet the panel is far from settled on where it belongs. Per-model borda runs from 0 to 60 with a standard deviation of 20.58, which the report labels "models split" — mistral-medium-3.5 ranked it median 3 across all five runs, while claude-opus-5 and deepseek-v4.1-flash placed it at median rank 5 where they named it at all. The gap reflects a shared read applied with different weight rather than genuine disagreement about the product: every model describes the R1T and R1S as capable, adventure-focused vehicles with strong software, then discounts them for premium pricing, a thin service network, and the financial and reliability uncertainty of a young automaker.

Rivian

per-model scores 060 of 100 · mean 25 across 6 models

The evidence sits under a consistent set of automotive review sources, with MotorTrendwww.motortrend.com×21 and Edmundswww.edmunds.com×29 recurring across models, alongside Consumer Reports where reliability risk is raised. What separates the rankings is how far each model lets the caveats pull the brand down: for the more cautious models the young-company risk and thin support network reduce it to a niche recommendation, while mistral-medium-3.5 treats the same limitations as constraints on breadth rather than on quality. No model ranked it first in any run.

The R1T and R1S are arguably the best-driving adventure EVs available, with excellent software, clever packaging, and some of the highest owner-satisfaction scores in the industry.

kimi-k3

Rivian excels in adventure-focused EVs with impressive off-road capabilities and a strong brand identity.

mistral-medium-3.5

Per model summary

  • mistral-medium-3.5
    brand score
    60
    first choice
    33
    median rank
    3.0
    present in
    5/5

    Rivian excels in adventure-focused EVs with off-road capability and premium build quality, though a limited model lineup, higher prices, and still-growing production scale restrict broader appeal.

  • gemini-3.7-flash
    brand score
    40
    first choice
    26
    median rank
    4.0
    present in
    5/5

    Rivian excels in premium, adventure-focused trucks and SUVs with off-road capability, refined interiors, and strong software, but higher prices and a still-expanding service and charging network limit its appeal for everyday drivers.

  • kimi-k3
    brand score
    28
    first choice
    19
    median rank
    4.0
    present in
    4/5

    The R1T and R1S are described as outstanding adventure EVs with clever design and high owner satisfaction, but high prices, a thin service network, and financial/reliability risks for a young company make it a compelling yet riskier choice.

  • claude-opus-5
    brand score
    8
    first choice
    8
    median rank
    5.0
    present in
    2/5

    Rivian's R1T and R1S are praised as capable, well-engineered adventure vehicles with strong software and loyal owners, but the recommendation is qualified by premium pricing, thin service networks, and young-automaker financial and reliability uncertainty.

  • deepseek-v4.1-flash
    brand score
    12
    first choice
    12
    median rank
    5.0
    present in
    3/5

    The R1T and R1S are praised for off-road capability and performance, but high prices and a still-growing service network make Rivian a niche choice for adventure enthusiasts rather than most buyers.

Sources per model

Each model’s own references for this brand, grouped by site. Tile area is that site’s share of the model’s references; tap one for the pages behind it.

mistral-medium-3.55 sites · 15 references
Rivian Official Site×5 · 33%

rivian.com

gemini-3.7-flash8 sites · 15 references
MotorTrend×4 · 27%

motortrend.com

kimi-k36 sites · 12 references
Consumer Reports×3 · 25%

consumerreports.org

claude-opus-53 sites · 6 references
Consumer Reports×2 · 33%

consumerreports.org

  • ×1/
  • ×1Consumer Reports reliability rankings /cars
deepseek-v4.1-flash4 sites · 9 references
Car and Driver×3 · 33%

caranddriver.com

Sources are what each model recalled as having shaped its view — not verified citations. A model without web access reconstructs a reference from memory, so a link may not lead where the model thought it did.

Category · EV vehicles
#5

BMW

22
brand score
14first choice
12345

Summary

BMW lands at a Borda score of 22 of 100, held down by being ranked by only 4 of the 6 models rather than by any hostility from those that did name it. Among the four that ranked it, placement is strikingly stable: gemini-3.7-flash gave it 44, claude-opus-5 and gpt-6-astra both 40, with each of those three covering it in every run and consistently settling on a median rank of 4. The gap that the report labels "models split" comes almost entirely from the two silent models scoring 0 and from kimi-k3's thin 20% coverage yielding a borda of 8.

BMW

per-model scores 044 of 100 · mean 22 across 6 models

The themes are consistent across the models that engaged with it: refined driving dynamics, quiet and well-built premium cabins, and above-average reliability, offset uniformly by higher purchase prices, costly options, and 400V-class charging that trails the 800V Korean rivals. That is why the brand clusters at fourth rather than higher — the models frame the ranking as a value trade-off, not an EV weakness. The recalled sources reflect this enthusiast-and-ownership lens, with Car and Driver×10 and Edmunds×6 recurring alongside Consumer Reports×7 on the reliability point.

The models agree BMW is a strong premium pick whose price, not its engineering, keeps it at fourth.

BMW is my strongest premium pick here for refined cabins, ride quality, and engaging driving dynamics across its electric range.

gpt-6-astra

They rank lower mainly on value: prices are high, options add up quickly, and charging speeds are good rather than class-leading.

kimi-k3

Per model summary

  • claude-opus-5
    brand score
    40
    first choice
    25
    median rank
    4.0
    present in
    5/5

    Consistently frames BMW's i4/iX/i5 as refined, well-built premium EVs with above-average reliability and traditional luxury feel, while noting higher costs and slower charging than 800V Korean rivals.

  • gemini-3.7-flash
    brand score
    44
    first choice
    27
    median rank
    4.0
    present in
    5/5

    Repeatedly emphasizes that BMW translates its signature driving dynamics and luxury craftsmanship into EVs with quiet, refined cabins, but cites higher prices and lower efficiency/slower charging as drawbacks.

  • gpt-6-astra
    brand score
    40
    first choice
    25
    median rank
    4.0
    present in
    5/5

    Consistently positions BMW as a premium pick for refinement, cabin quality, and engaging driving dynamics, ranking it fourth mainly due to high purchase prices and costly options rather than any EV weakness.

  • kimi-k3
    brand score
    8
    first choice
    5
    median rank
    4.0
    present in
    1/5

    Describes BMW's i4 and iX as refined, well-built premium EVs with respectable range, ranking lower on value due to high prices, costly options, and merely good charging speeds.

Sources per model

Each model’s own references for this brand, grouped by site. Tile area is that site’s share of the model’s references; tap one for the pages behind it.

claude-opus-56 sites · 15 references
Car and Driver×5 · 33%

caranddriver.com

gemini-3.7-flash7 sites · 15 references
Car and Driver×4 · 27%

caranddriver.com

gpt-6-astra2 sites · 10 references
Car and Driver×5 · 50%

caranddriver.com

kimi-k33 sites · 3 references
Car and Driver×1 · 33%

caranddriver.com

Sources are what each model recalled as having shaped its view — not verified citations. A model without web access reconstructs a reference from memory, so a link may not lead where the model thought it did.

Category · EV vehicles

How this was measured

The question

  • Unaided brand recommendation question: “Which EV brand would you recommend?
  • The prompt asks each model to return exactly 5 brands, ranked 1 to 5, and for each one a reason for the recommendation and the sources that informed it.

Sampling

  • Models are not deterministic, so the question is asked over and over — 5 answers per model
  • Spellings of the same brand are normalized to the most commonly used form before counting

Measures of Position: Median and Quartiles

  • Median value indicates that in 50% of answers the brand held this position or higher.
  • Quartiles help visualize the spread of rankings per brand. Q1 indicates that 25% of answers had this rank or higher. Q3 means that 75% of answers ranked the brand as X or better.
  • Medians and quartiles are computed only from answers where the brand was present.

Coverage

  • Count of answers in which the given brand was named, as a percentage.

Brand score - Normalized Borda score

  • The brand score is calculated using a normalized Borda score. Borda count is a voting method: each ballot awards points by position instead of naming one winner. Our ranking responses from the LLM are always fixed to 5 answers, so we assign 100 points for rank 1, then 80, 60, 40, 20 — and 0 if the brand is missing.
Brand score=100ni=1n6ri5\text{Brand\ score} = \frac{100}{n} \sum_{i=1}^{n} \frac{6 - r_i}{5}

where

ri={1,,5if the brand appears6if the brand is not mentionedr_i = \begin{cases} 1,\dots,5 & \text{if the brand appears} \\ 6 & \text{if the brand is not mentioned} \end{cases}
  • This way we can calculate a common score for all brands mentioned across model responses. The score reflects both how high the brand was ranked and how frequently it was mentioned.

First choice score - adjusted MRR - top-rank indicator

  • The first choice score is based on Mean Reciprocal Rank (a common search-engine measure), but adapted to measure the position of a specific brand across repeated ranked responses.
  • Each occurrence receives a reciprocal-position score: 100 for 1st, 50 for 2nd, 33.3 for 3rd, 25 for 4th, 20 for 5th, and 0 when the brand is not mentioned. The scores are averaged across all responses.
First choice score=100ni=1nsi\text{First\ choice\ score} = \frac{100}{n} \sum_{i=1}^{n} s_i

where

si={1/riif the brand is present0if the brand is absents_i = \begin{cases} 1/r_i & \text{if the brand is present} \\ 0 & \text{if the brand is absent} \end{cases}
  • No rank below 1st place gets more than 50, so the score is heavily driven by first places.
Your category

Ask your own question of every model

Tell us what to ask and how wide to sample it — we run it and send back the report.

Sample size
Display results
0/1000

Sampling every model takes up to 24 hours once we start.