Customer data platforms
Asked as “customer data platform”, and as “CDP”, on behalf of a mid-market B2B company. 49 first choices recorded across the direct, paraphrase, budget and scale prompts, fourteen models each.
Standing · first-choice share
49%
Clear leader
49Segment20Hightouch08BlueConic22others
49% of first choices, clear leader.
Since September 2026▲+3Since September 2026: 48% → 51%, +3 points. Inside the 11-point floor: within noise. Read over the models both editions asked.Segment held the lead, +3 points on 48%, inside the 11-point floor.
By buyer segment
The same question asked on behalf of a different buyer. Each standing is computed within its segment; they sit side by side and are never added together.
Small business
Segment35%Contested▼−3Since September 2026: 39% → 36%, −3 points. Inside the 11-point floor: within noise. Read over the models both editions asked.Standing →Mid-marketSegment49%Clear leader▲+3Since September 2026: 48% → 51%, +3 points. Inside the 11-point floor: within noise. Read over the models both editions asked.This pageEnterprise
Salesforce Data Cloud34%Contested▼−7Since September 2026: 41% → 34%, −7 points. Inside the 11-point floor: within noise. Read over the models both editions asked.Standing →
Segment35%Contested▼−3Since September 2026: 39% → 36%, −3 points. Inside the 11-point floor: within noise. Read over the models both editions asked.Standing →Mid-marketSegment49%Clear leader▲+3Since September 2026: 48% → 51%, +3 points. Inside the 11-point floor: within noise. Read over the models both editions asked.This pageEnterpriseSalesforce Data Cloud34%Contested▼−7Since September 2026: 41% → 34%, −7 points. Inside the 11-point floor: within noise. Read over the models both editions asked.Standing →The standing
Share is the count of first choices across the direct, paraphrase, budget and scale prompts, over all fourteen models, for a mid-market B2B company. Ordered by share.
ProductFirst-choice shareNegative rateLabelsQuadrantSince September 2026
01Segment49%14%64endorsed leader▲+3Since September 2026: 48% → 51%, +3 points. Inside the 11-point floor: within noise. Read over the models both editions asked.48% → 51%
Segment49%14%64endorsed leader▲+3Since September 2026: 48% → 51%, +3 points. Inside the 11-point floor: within noise. Read over the models both editions asked.48% → 51%02
Hightouch20%5%42accepted challenger▼−1Since September 2026: 20% → 20%, −1 point. Inside the 11-point floor: within noise. Read over the models both editions asked.20% → 20%
Hightouch20%5%42accepted challenger▼−1Since September 2026: 20% → 20%, −1 point. Inside the 11-point floor: within noise. Read over the models both editions asked.20% → 20%03
BlueConic8%15%20accepted challenger=heldSince September 2026: 5% → 5%, ±0 points. Inside the 11-point floor: within noise. Read over the models both editions asked.5% → 5%
BlueConic8%15%20accepted challenger=heldSince September 2026: 5% → 5%, ±0 points. Inside the 11-point floor: within noise. Read over the models both editions asked.5% → 5%04
RudderStack4%0%31accepted challenger▼−11Since September 2026: 14% → 2%, −11 points. Past the 11-point floor: movement. Read over the models both editions asked.14% → 2%
RudderStack4%0%31accepted challenger▼−11Since September 2026: 14% → 2%, −11 points. Past the 11-point floor: movement. Read over the models both editions asked.14% → 2%05
Census2%0%11accepted challenger▲+2Since September 2026: 0% → 2%, +2 points. Inside the 11-point floor: within noise. Read over the models both editions asked.0% → 2%
Census2%0%11accepted challenger▲+2Since September 2026: 0% → 2%, +2 points. Inside the 11-point floor: within noise. Read over the models both editions asked.0% → 2%06Salesforce Data Cloud2%38%37criticized challenger▲+2Since September 2026: 0% → 2%, +2 points. Inside the 11-point floor: within noise. Read over the models both editions asked.0% → 2%
Salesforce Data Cloud2%38%37criticized challenger▲+2Since September 2026: 0% → 2%, +2 points. Inside the 11-point floor: within noise. Read over the models both editions asked.0% → 2%07
Adobe Real-Time CDP2%50%34criticized challenger▲+2Since September 2026: 0% → 2%, +2 points. Inside the 11-point floor: within noise. Read over the models both editions asked.0% → 2%
Adobe Real-Time CDP2%50%34criticized challenger▲+2Since September 2026: 0% → 2%, +2 points. Inside the 11-point floor: within noise. Read over the models both editions asked.0% → 2%Show the five products at 0%, ordered by negative rate
08
Tealium0%21%38accepted challenger=heldSince September 2026: 0% → 0%, ±0 points. Inside the 11-point floor: within noise. Read over the models both editions asked.0% → 0%
Tealium0%21%38accepted challenger=heldSince September 2026: 0% → 0%, ±0 points. Inside the 11-point floor: within noise. Read over the models both editions asked.0% → 0%10
mParticle0%20%20accepted challenger=heldSince September 2026: 0% → 0%, ±0 points. Inside the 11-point floor: within noise. Read over the models both editions asked.0% → 0%
mParticle0%20%20accepted challenger=heldSince September 2026: 0% → 0%, ±0 points. Inside the 11-point floor: within noise. Read over the models both editions asked.0% → 0%11
Klaviyo Data Platform0%17%12accepted challenger=heldSince September 2026: 0% → 0%, ±0 points. Inside the 11-point floor: within noise. Read over the models both editions asked.0% → 0%
Klaviyo Data Platform0%17%12accepted challenger=heldSince September 2026: 0% → 0%, ±0 points. Inside the 11-point floor: within noise. Read over the models both editions asked.0% → 0%12
ActionIQ0%15%13accepted challenger=heldSince September 2026: 0% → 0%, ±0 points. Inside the 11-point floor: within noise. Read over the models both editions asked.0% → 0%
ActionIQ0%15%13accepted challenger=heldSince September 2026: 0% → 0%, ±0 points. Inside the 11-point floor: within noise. Read over the models both editions asked.0% → 0%09
Treasure Data0%7%15accepted challenger=heldSince September 2026: 0% → 0%, ±0 points. Inside the 11-point floor: within noise. Read over the models both editions asked.0% → 0%
Treasure Data0%7%15accepted challenger=heldSince September 2026: 0% → 0%, ±0 points. Inside the 11-point floor: within noise. Read over the models both editions asked.0% → 0%The floor is 11 points of share, measured: how far the models move a leader on their own when the same questions are asked twice with nothing changed. A larger change is movement; a smaller one is noise, and both are shown. Movement is read over the twelve models both editions asked; GPT-6 Luna, Muse Glimmer 30B joined this edition and are in the standing but not yet in the comparison. How the floor is measured
Bars are the share of first choices, 0 to 100Every product with at least 10 labels here. Every product name links to its product page.
All twenty-one head-to-head pages: the top seven products, each against each
Segment vs CensusSegment vs Salesforce Data CloudSegment vs Adobe Real-Time CDPHightouch vs BlueConicHightouch vs RudderStackHightouch vs CensusHightouch vs Salesforce Data CloudHightouch vs Adobe Real-Time CDPBlueConic vs RudderStackBlueConic vs CensusBlueConic vs Salesforce Data CloudBlueConic vs Adobe Real-Time CDPRudderStack vs CensusRudderStack vs Salesforce Data CloudRudderStack vs Adobe Real-Time CDPCensus vs Salesforce Data CloudCensus vs Adobe Real-Time CDPSalesforce Data Cloud vs Adobe Real-Time CDP
Recommended versus criticized
Every product with at least 10 labels here, on both axes. The 30% line names a quadrant, not the verdict above: that one needs more than 40%.
Criticized challengerCriticized default
Negative label rate →
010203040506070809101112Accepted challengerEndorsed leader
0%First-choice share → · lines at 30% share and 25% negative60%
Key
01Segment49%
Segment49%02Hightouch20%
Hightouch20%03BlueConic8%
BlueConic8%04RudderStack4%
RudderStack4%05Census2%
Census2%06Salesforce Data Cloud2%
Salesforce Data Cloud2%07Adobe Real-Time CDP2%
Adobe Real-Time CDP2%08Tealium0%
Tealium0%09Treasure Data0%
Treasure Data0%10mParticle0%
mParticle0%11Klaviyo Data Platform0%
Klaviyo Data Platform0%12ActionIQ0%
ActionIQ0%What they warned about
Six of fourteen models held their first choice under the paraphrase. Claude Haiku 4.5, Gemini 3.5 Flash, Mistral Small, DeepSeek V4 Flash, Llama 4 Maverick, Qwen 3.7 Flash, Kimi K2 and MiniMax M2.5 changed. A high negative share on a product with few labels is a warning. A low share on a product with many labels is salience, not sentiment.
Segment14%
9 of 64 labels negative · 8 of 14 models
“Segment's documentation says some deletion requests depend on an identifiable `userId`, are asynchronous... Those are specific limits to test—not a blanket reason to reject Segment.” GPT-6 Luna, negative prompt
Adobe Real-Time CDP50%
17 of 34 labels negative · 7 of 14 models · 2 hard negative
“You should **avoid** these if you do not have an enterprise budget or IT headcount ... they carry a premium price tag and high complexity” Qwen 3.7 Flash, negative prompt
Salesforce Data Cloud38%
14 of 37 labels negative · 7 of 14 models · 2 hard negative
“You should **avoid** these if you do not have an enterprise budget or IT headcount ... premium price tag and high complexity” Qwen 3.7 Flash, negative prompt
Tealium21%
8 of 38 labels negative · 6 of 14 models · 2 hard negative
“What to avoid on a tight budget * Tealium publishes no pricing; Vendr pegs mid-market deployments at $60,000–$150,000 per year” Muse Glimmer 30B, budget prompt
What they cite
Citations exist only for the models that return a source list: fourteen of the fourteen in this edition, and all six flagship models on the expanded tier.
Sites the answers cite
75 of 84 answers in this category came back with a source list, from 14 of 14 models: citations where the model returns them, or the search results it consulted. 1003 links across 245 sites, every framing counted. Ranked by the number of answers carrying the site or page. 9 of the 252 answers across every segment cited this index's own page for the category; the method page measures whether that reading tilts an answer.
52 answers · 84 citations · 13 models
27 answers · 50 citations · 12 models
vendor site · Improvado23 answers · 25 citations · 11 models
vendor site · Gartner22 answers · 31 citations · 12 models
vendor site · Brevo22 answers · 22 citations · 7 models
vendor site · Rework21 answers · 27 citations · 9 models
20 answers · 20 citations · 7 models
19 answers · 20 citations · 9 models
vendor site · Hightouch18 answers · 24 citations · 8 models
18 answers · 18 citations · 10 models
vendor site · CloudTalk14 answers · 14 citations · 7 models
12 answers · 23 citations · 8 models
Pages the answers cite
The ten pages named in the most answers, by full address. A page here is one the models returned with a recommendation, not one the index endorses.
43 answers · 12 models
23 answers · 11 models
22 answers · 7 models
20 answers · 7 models
19 answers · 9 models
19 answers · 11 models
18 answers · 10 models
14 answers · 7 models
11 answers · 7 models
11 answers · 7 models
Search against answers
Each company's standing in the answers beside its site's footprint in Google search, one row a site: the products the models named on it with their shares, and the share they add up to; monthly searches on Google, and DataForSEO's estimate of AI search demand (modeled from search signals, directional, not a count of queries to any assistant), for the most-searched of the company's and its products' names (the name is in each row's hover text); estimated monthly organic visits to the site; and its best position in Google's top ten for “best customer data platform”, “customer data platform”. US estimates from DataForSEO and Google's Ads Transparency Center. A small company's site, or a mid-sized company's site for its flagship, is marked company; a product on a large parent's site (Google, Microsoft) has no site figures. A column with no figures for this category is left out, and an empty cell means none were seen, not none exist. Two measurements side by side: neither is read as the cause of the other.
| Company and products | Share | Own site cited | Name searches, Google | AI search demand, est. | Organic visits | Paid search a month, est. | Google ads, last 30 days | Best Google position |
|---|---|---|---|---|---|---|---|---|
| Twilio segment.com Segment 49% | 49% | 408 | 135,000 | 4,343 | 9,086 | 0 | ||
| Hightouch hightouch.com | 20% | 363 | 12,100 | 434 | 127,772 | 23 | 7 | |
| BlueConic blueconic.com | 8% | 64 | 1,000 | 22 | 4,661 | $142 | 9 | 1 |
| RudderStack rudderstack.com | 4% | 125 | 1,900 | 42 | 46,132 | 50 | ||
| Able CDP ablecdp.com | 4% | 4 | 30 | 0 | 172 | |||
| Fivetran fivetran.com Census 2% | 2% | 224 | 60,500 | 14,424 | 41,501 company | $126 | 97 | |
| Salesforce parent site Salesforce Data Cloud 2% | 2% | 320 | 550,000 | 60,632 | ||||
| Brevo brevo.com | 2% | 386 | 40,500 | 327 | 81,741 | $17,847 | 106 | |
| Customer.io customer.io | 2% | 187 | 2,900 | 174 | 15,750 | $6,920 | 112 | |
| ZoomInfo zoominfo.com | 2% | 846 | 135,000 | 1,810 | 6,376,738 | $1,140 | 120 |
Google's top ten for the two searches hold 16 sites; 5 of them are among the sites the models cited here (cdp.com, cxtoday.com, gartner.com, hightouch.com, learn.g2.com). The first result for “best customer data platform” is gartner.com.
How they answered
Six framings of the same buying question, each sent to every model in a fresh session with search on. One row per model, so a row shows whether it held its answer under rewording, what it named when cost was the constraint, and what it argued against. Computed from the raw judge labels.ShowHide
How they answered
Six framings of the same buying question, each sent to every model in a fresh session with search on. One row per model, so a row shows whether it held its answer under rewording, what it named when cost was the constraint, and what it argued against. Computed from the raw judge labels.
ShowHideModelDirect“What is the best customer data platform for a mid-market B2B company?”Paraphrase“Which CDP would you recommend to a mid-sized B2B company?”Comparative“What are the top customer data platforms and how do they differ?”Budget-constrained“What is the best customer data platform for a company with a limited budget?”Scale-constrained“We are a 500 person company evaluating a customer data platform. What should we look at?”Negative“Which customer data platforms should I avoid or be cautious about?”
Claude Haiku 4.5BlueConic, Segmentno first choiceChangedno first choiceSegmentno first choicenothing named
Two alternatives
Hightouch, Salesforce Data CloudThree alternatives
Hightouch, Klaviyo Data Platform, RudderStackGPT-5.4 miniHightouchHightouchHeldno first choiceSegmentno first choicenothing named
Two alternatives
RudderStack, SegmentTwo alternatives
RudderStack, mParticleThree alternatives
Brevo, CustomerLabs, OmnisendGemini 3.5 FlashHightouchagainst: Adobe Real-Time CDP, Salesforce Data CloudCensus, HightouchChangedagainst: Adobe Real-Time CDP, Salesforce Data Cloudno first choiceHightouchagainst: Adobe Real-Time CDP, Salesforce Data Cloud, Tealiumno first choiceagainst: Adobe Real-Time CDP, Salesforce Data Cloudagainst: Adobe Real-Time CDP, BlueConic, Salesforce Data Cloud, Segment, Tealium, mParticle
Three alternatives
BlueConic, CaliberMind, SegmentThree alternatives
BlueConic, HubSpot Breeze Intelligence, SegmentNine alternatives
ActionIQ, Adobe Real-Time CDP, Bloomreach, GrowthLoop, Hightouch, Insider, Salesforce Data Cloud, Segment, TealiumFive alternatives
Able CDP, Customer.io, Jitsu, Klaviyo Data Platform, RudderStackPerplexity SonarSegmentSegmentHeldagainst: Adobe Real-Time CDPno first choiceBrevono first choiceagainst: Adobe Real-Time CDP, BlueConic, Hightouch, Segment, Simon Data, Tealium, mParticle
Three alternatives
BlueConic, Hightouch, TealiumOne alternative
HightouchTwo alternatives
RudderStack, SegmentGrok 4.1 FastSegmentSegmentHeldno first choiceSegmentagainst: Adobe Real-Time CDP, Salesforce Data Cloudno first choiceagainst: Adobe Real-Time CDPagainst: Adobe Real-Time CDP, Hightouch, Klaviyo Data Platform, Salesforce Data Cloud, Segment
Two alternatives
BlueConic, HightouchOne alternative
BlueConicThree alternatives
Census, Hightouch, RudderStackMistral SmallHightouch, SegmentAdobe Real-Time CDP, Salesforce Data CloudChangedSegmentCustomer.io, Segmentno first choicenothing named
Three alternatives
BlueConic, Demandbase, TealiumTwo alternatives
Microsoft Customer Data Platform, SALESmanagoFive alternatives
ActionIQ, Adobe Real-Time CDP, Bloomreach, Salesforce Data Cloud, TealiumOne alternative
PipedriveDeepSeek V4 FlashHightouch, SegmentSegmentChangedagainst: Salesforce Data CloudSegmentagainst: Klaviyo Data PlatformAble CDPagainst: ActionIQ, Adobe Real-Time CDP, Salesforce Data Cloud, Tealiumno first choiceagainst: Adobe Real-Time CDP, Segment, Tealium, mParticle
Two alternatives
BlueConic, RudderStackTwo alternatives
BlueConic, HightouchNine alternatives
Adobe Real-Time CDP, Amperity, BlueConic, Hightouch, RudderStack, Salesforce Data Cloud, Tealium, Treasure Data, mParticleThree alternatives
Klaviyo Data Platform, RudderStack, SegmentLlama 4 MaverickZoomInfono first choiceChangedno first choiceno first choiceno first choicenothing named
Three alternatives
Hightouch, Segment, TealiumQwen 3.7 FlashHightouch, Segmentagainst: Salesforce Data CloudSegmentChangedno first choiceagainst: Adobe Real-Time CDPAble CDPagainst: Adobe Real-Time CDP, Segment, Tealiumno first choiceagainst: Acquia, Adobe Real-Time CDP, Amperity, SAP Customer Data Platform, Salesforce Data Cloud
Four alternatives
AudienceStream, HubSpot, RudderStack, mParticleTwo alternatives
BlueConic, SALESmanagoFive alternatives
Amperity, Hightouch, Salesforce Data Cloud, Segment, TealiumSix alternatives
Census, Hightouch, HubSpot, Jitsu, PostHog, RudderStackKimi K2BlueConicagainst: SegmentSegmentChangedno first choiceRudderStack, Segmentno first choiceagainst: Adobe Real-Time CDP, Salesforce Data Cloud, Segment, Tealium, Treasure Data, mParticle
Two alternatives
Salesforce Data Cloud, TealiumTwo alternatives
HubSpot Operations Hub, Treasure DataTwo alternatives
Hightouch, Klaviyo Data PlatformFour alternatives
Hightouch, Segment, Tealium, mParticleGLM 4.7 FlashXSegmentSegmentHeldno first choiceSegmentno first choicenothing named
Three alternatives
6sense, HubSpot, Salesforce Data CloudTwo alternatives
Hightouch, Salesforce Data CloudSeven alternatives
ActionIQ, Adobe Real-Time CDP, Hightouch, Salesforce Data Cloud, Segment, Tealium, Treasure DataFive alternatives
AtroCore, Coffee, PostHog, RudderStack, TracardiMiniMax M2.5HubSpot CDPSegmentChangedSegmentSegmentagainst: Adobe Real-Time CDP, Salesforce Data Cloudno first choicenothing named
Four alternatives
Bloomreach, RudderStack, Segment, TealiumThree alternatives
Bloomreach, RudderStack, TealiumSix alternatives
Adobe Real-Time CDP, Hightouch, RudderStack, Salesforce Data Cloud, Tealium, Treasure DataFour alternatives
Brevo, Customer.io, HubSpot CRM, Klaviyo Data PlatformGPT-6 LunaHightouchHightouchHeldno first choiceRudderStackno first choiceagainst: Segment
Two alternatives
RudderStack, SegmentOne alternative
SegmentEight alternatives
Adobe Real-Time CDP, Hightouch, RudderStack, Salesforce Data Cloud, Segment, Tealium, Treasure Data, mParticleTwo alternatives
PostHog, SegmentMuse Glimmer 30BBlueConic, SegmentBlueConic, SegmentHeldagainst: ActionIQno first choiceagainst: BlueConicSegmentagainst: Tealiumno first choiceagainst: Salesforce Data Cloud, Segment
Four alternatives
Hightouch, Salesforce Data Cloud, Tealium, Treasure DataFour alternatives
Hightouch, RudderStack, Salesforce Data Cloud, TealiumTwo alternatives
Salesforce Data Cloud, TealiumThree alternatives
BlueConic, HubSpot, RudderStackBold is the first choiceAlternatives are counted; the count opens them.What the answer argued against
The record
One row per call: the version string exactly as returned, whether the model searched, sources cited, and latency. Full answer text is in the free responses file. Download the record
Eighty-four rows: every prompt, every model, every answer.
| Prompt | Model | Version string | Time (UTC) | Searched | Sources | Latency |
|---|---|---|---|---|---|---|
| Direct recommendation | Claude Haiku 4.5 | claude-haiku-4-5-20251001 | 2026-10-01 10:00 | yes | 9 | 8 s |
| Direct recommendation | GPT-5.4 mini | gpt-5.4-mini-2026-03-17 | 2026-10-01 11:58 | yes | 3 | 6 s |
| Direct recommendation | Gemini 3.5 Flash | gemini-3.5-flash | 2026-10-01 08:54 | yes | 22 | 27 s |
| Direct recommendation | Perplexity Sonar | sonar | 2026-10-01 12:28 | yes | 20 | 3 s |
| Direct recommendation | Grok 4.1 Fast | spacexai/grok-4.1-fast-non-reasoning via vertex | 2026-10-01 08:19 | yes | 22 | 7 s |
| Direct recommendation | Mistral Small | mistral/mistral-small via mistral | 2026-10-01 09:26 | yes | 5 | 3 s |
| Direct recommendation | DeepSeek V4 Flash | deepseek/deepseek-v4-flash via deepinfra | 2026-10-01 12:22 | yes | 17 | 23 s |
| Direct recommendation | Llama 4 Maverick | meta/llama-4-maverick via bedrock | 2026-10-01 14:11 | yes | 5 | 2 s |
| Direct recommendation | Qwen 3.7 Flash | alibaba/qwen3.7-flash via alibaba | 2026-10-01 09:33 | no | 0 | 39 s |
| Direct recommendation | Kimi K2 | moonshotai/kimi-k2 via novita | 2026-10-01 08:06 | yes | 14 | 21 s |
| Direct recommendation | GLM 4.7 FlashX | zai/glm-4.7-flashx via zai | 2026-10-01 12:58 | yes | 21 | 113 s |
| Direct recommendation | MiniMax M2.5 | minimax/minimax-m2.5 via minimax | 2026-10-01 09:42 | yes | 5 | 14 s |
| Direct recommendation | GPT-6 Luna | gpt-6-luna | 2026-10-01 10:04 | yes | 3 | 13 s |
| Direct recommendation | Muse Glimmer 30B | meta/muse-glimmer-30b via togetherai | 2026-10-01 11:15 | yes | 11 | 17 s |
| Paraphrase | Claude Haiku 4.5 | claude-haiku-4-5-20251001 | 2026-10-01 13:18 | no | 0 | 3 s |
| Paraphrase | GPT-5.4 mini | gpt-5.4-mini-2026-03-17 | 2026-10-01 14:01 | yes | 3 | 4 s |
| Paraphrase | Gemini 3.5 Flash | gemini-3.5-flash | 2026-10-01 09:35 | yes | 15 | 21 s |
| Paraphrase | Perplexity Sonar | sonar | 2026-10-01 08:17 | yes | 18 | 3 s |
| Paraphrase | Grok 4.1 Fast | spacexai/grok-4.1-fast-non-reasoning via vertex | 2026-10-01 13:18 | yes | 22 | 8 s |
| Paraphrase | Mistral Small | mistral/mistral-small via mistral | 2026-10-01 08:32 | yes | 25 | 8 s |
| Paraphrase | DeepSeek V4 Flash | deepseek/deepseek-v4-flash via deepinfra | 2026-10-01 13:03 | yes | 21 | 30 s |
| Paraphrase | Llama 4 Maverick | meta/llama-4-maverick via bedrock | 2026-10-01 11:33 | yes | 5 | 3 s |
| Paraphrase | Qwen 3.7 Flash | alibaba/qwen3.7-flash via alibaba | 2026-10-01 12:15 | yes | 7 | 36 s |
| Paraphrase | Kimi K2 | moonshotai/kimi-k2 via novita | 2026-10-01 09:46 | yes | 17 | 20 s |
| Paraphrase | GLM 4.7 FlashX | zai/glm-4.7-flashx via zai | 2026-10-01 11:47 | yes | 15 | 76 s |
| Paraphrase | MiniMax M2.5 | minimax/minimax-m2.5 via minimax | 2026-10-01 13:01 | no | 0 | 11 s |
| Paraphrase | GPT-6 Luna | gpt-6-luna | 2026-10-01 07:36 | yes | 2 | 16 s |
| Paraphrase | Muse Glimmer 30B | meta/muse-glimmer-30b via togetherai | 2026-10-01 08:40 | yes | 13 | 17 s |
| Comparative | Claude Haiku 4.5 | claude-haiku-4-5-20251001 | 2026-10-01 09:37 | yes | 9 | 6 s |
| Comparative | GPT-5.4 mini | gpt-5.4-mini-2026-03-17 | 2026-10-01 13:42 | yes | 5 | 9 s |
| Comparative | Gemini 3.5 Flash | gemini-3.5-flash | 2026-10-01 10:41 | yes | 17 | 23 s |
| Comparative | Perplexity Sonar | sonar | 2026-10-01 11:09 | yes | 20 | 6 s |
| Comparative | Grok 4.1 Fast | spacexai/grok-4.1-fast-non-reasoning via vertex | 2026-10-01 08:59 | yes | 21 | 8 s |
| Comparative | Mistral Small | mistral/mistral-small via mistral | 2026-10-01 07:27 | yes | 13 | 15 s |
| Comparative | DeepSeek V4 Flash | deepseek/deepseek-v4-flash via deepinfra | 2026-10-01 11:43 | yes | 24 | 44 s |
| Comparative | Llama 4 Maverick | meta/llama-4-maverick via bedrock | 2026-10-01 11:38 | yes | 5 | 3 s |
| Comparative | Qwen 3.7 Flash | alibaba/qwen3.7-flash via alibaba | 2026-10-01 11:06 | yes | 10 | 37 s |
| Comparative | Kimi K2 | moonshotai/kimi-k2 via novita | 2026-10-01 11:56 | yes | 12 | 25 s |
| Comparative | GLM 4.7 FlashX | zai/glm-4.7-flashx via zai | 2026-10-01 11:37 | yes | 16 | 25 s |
| Comparative | MiniMax M2.5 | minimax/minimax-m2.5 via minimax | 2026-10-01 08:36 | yes | 9 | 22 s |
| Comparative | GPT-6 Luna | gpt-6-luna | 2026-10-01 11:38 | yes | 10 | 20 s |
| Comparative | Muse Glimmer 30B | meta/muse-glimmer-30b via togetherai | 2026-10-01 09:02 | yes | 13 | 23 s |
| Budget constrained | Claude Haiku 4.5 | claude-haiku-4-5-20251001 | 2026-10-01 11:56 | yes | 9 | 7 s |
| Budget constrained | GPT-5.4 mini | gpt-5.4-mini-2026-03-17 | 2026-10-01 12:52 | yes | 3 | 5 s |
| Budget constrained | Gemini 3.5 Flash | gemini-3.5-flash | 2026-10-01 10:38 | yes | 28 | 25 s |
| Budget constrained | Perplexity Sonar | sonar | 2026-10-01 10:07 | yes | 21 | 3 s |
| Budget constrained | Grok 4.1 Fast | spacexai/grok-4.1-fast-non-reasoning via vertex | 2026-10-01 09:41 | yes | 23 | 8 s |
| Budget constrained | Mistral Small | mistral/mistral-small via mistral | 2026-10-01 10:02 | yes | 5 | 4 s |
| Budget constrained | DeepSeek V4 Flash | deepseek/deepseek-v4-flash via deepinfra | 2026-10-01 13:53 | yes | 21 | 37 s |
| Budget constrained | Llama 4 Maverick | meta/llama-4-maverick via bedrock | 2026-10-01 10:09 | yes | 5 | 1 s |
| Budget constrained | Qwen 3.7 Flash | alibaba/qwen3.7-flash via alibaba | 2026-10-01 11:01 | yes | 9 | 46 s |
| Budget constrained | Kimi K2 | moonshotai/kimi-k2 via novita | 2026-10-01 10:44 | yes | 14 | 16 s |
| Budget constrained | GLM 4.7 FlashX | zai/glm-4.7-flashx via zai | 2026-10-01 14:09 | yes | 14 | 112 s |
| Budget constrained | MiniMax M2.5 | minimax/minimax-m2.5 via minimax | 2026-10-01 11:16 | yes | 10 | 20 s |
| Budget constrained | GPT-6 Luna | gpt-6-luna | 2026-10-01 08:52 | yes | 3 | 10 s |
| Budget constrained | Muse Glimmer 30B | meta/muse-glimmer-30b via togetherai | 2026-10-01 14:12 | yes | 14 | 16 s |
| Scale constrained | Claude Haiku 4.5 | claude-haiku-4-5-20251001 | 2026-10-01 10:05 | no | 0 | 6 s |
| Scale constrained | GPT-5.4 mini | gpt-5.4-mini-2026-03-17 | 2026-10-01 10:16 | no | 0 | 7 s |
| Scale constrained | Gemini 3.5 Flash | gemini-3.5-flash | 2026-10-01 13:55 | yes | 5 | 26 s |
| Scale constrained | Perplexity Sonar | sonar | 2026-10-01 12:17 | yes | 16 | 5 s |
| Scale constrained | Grok 4.1 Fast | spacexai/grok-4.1-fast-non-reasoning via vertex | 2026-10-01 10:09 | yes | 14 | 8 s |
| Scale constrained | Mistral Small | mistral/mistral-small via mistral | 2026-10-01 09:27 | no | 0 | 7 s |
| Scale constrained | DeepSeek V4 Flash | deepseek/deepseek-v4-flash via deepinfra | 2026-10-01 13:35 | no | 0 | 18 s |
| Scale constrained | Llama 4 Maverick | meta/llama-4-maverick via bedrock | 2026-10-01 13:42 | yes | 5 | 2 s |
| Scale constrained | Qwen 3.7 Flash | alibaba/qwen3.7-flash via alibaba | 2026-10-01 09:00 | no | 0 | 40 s |
| Scale constrained | Kimi K2 | moonshotai/kimi-k2 via novita | 2026-10-01 12:38 | yes | 19 | 28 s |
| Scale constrained | GLM 4.7 FlashX | zai/glm-4.7-flashx via zai | 2026-10-01 08:31 | yes | 25 | 62 s |
| Scale constrained | MiniMax M2.5 | minimax/minimax-m2.5 via minimax | 2026-10-01 12:59 | no | 0 | 10 s |
| Scale constrained | GPT-6 Luna | gpt-6-luna | 2026-10-01 13:05 | yes | 2 | 22 s |
| Scale constrained | Muse Glimmer 30B | meta/muse-glimmer-30b via togetherai | 2026-10-01 09:36 | yes | 14 | 22 s |
| Negative framing | Claude Haiku 4.5 | claude-haiku-4-5-20251001 | 2026-10-01 12:29 | yes | 18 | 9 s |
| Negative framing | GPT-5.4 mini | gpt-5.4-mini-2026-03-17 | 2026-10-01 11:53 | yes | 5 | 7 s |
| Negative framing | Gemini 3.5 Flash | gemini-3.5-flash | 2026-10-01 10:41 | yes | 19 | 27 s |
| Negative framing | Perplexity Sonar | sonar | 2026-10-01 10:48 | yes | 19 | 4 s |
| Negative framing | Grok 4.1 Fast | spacexai/grok-4.1-fast-non-reasoning via vertex | 2026-10-01 08:49 | yes | 24 | 8 s |
| Negative framing | Mistral Small | mistral/mistral-small via mistral | 2026-10-01 12:47 | yes | 5 | 36 s |
| Negative framing | DeepSeek V4 Flash | deepseek/deepseek-v4-flash via deepinfra | 2026-10-01 10:49 | yes | 25 | 34 s |
| Negative framing | Llama 4 Maverick | meta/llama-4-maverick via bedrock | 2026-10-01 11:53 | yes | 5 | 4 s |
| Negative framing | Qwen 3.7 Flash | alibaba/qwen3.7-flash via alibaba | 2026-10-01 07:44 | yes | 10 | 31 s |
| Negative framing | Kimi K2 | moonshotai/kimi-k2 via novita | 2026-10-01 12:07 | yes | 25 | 47 s |
| Negative framing | GLM 4.7 FlashX | zai/glm-4.7-flashx via zai | 2026-10-01 13:41 | yes | 23 | 124 s |
| Negative framing | MiniMax M2.5 | minimax/minimax-m2.5 via minimax | 2026-10-01 12:41 | yes | 15 | 20 s |
| Negative framing | GPT-6 Luna | gpt-6-luna | 2026-10-01 12:16 | yes | 2 | 12 s |
| Negative framing | Muse Glimmer 30B | meta/muse-glimmer-30b via togetherai | 2026-10-01 08:16 | yes | 25 | 35 s |
Noise floor in this category
Flips between the edition run and its calibration repeat. Six prompts per model is a small sample; the index-wide floor is the number to trust.
ShowHide
Claude Haiku 4.5
1 of 3 flipped
GPT-5.4 mini
1 of 3 flipped
Gemini 3.5 Flash
3 of 4 flipped
Perplexity Sonar
1 of 3 flipped
Grok 4.1 Fast
2 of 4 flipped
Mistral Small
3 of 4 flipped
DeepSeek V4 Flash
4 of 5 flipped
Llama 4 Maverick
3 of 3 flipped
Qwen 3.7 Flash
2 of 3 flipped
Kimi K2
2 of 3 flipped
GLM 4.7 FlashX
2 of 3 flipped
MiniMax M2.5
3 of 4 flipped
GPT-6 Luna
0 of 3 flipped
Muse Glimmer 30B
1 of 4 flipped
Normalization in this category
Every judgment call made between the raw labels and the numbers above, listed so it is visible and reversible.
ShowHide
Category-scoped readings
Adobe read as Adobe Real-Time CDP
HubSpot read as HubSpot
HubSpot (Starter/Brownie Tier) read as HubSpot
Klaviyo read as Klaviyo Data Platform
Klaviyo CDP read as Klaviyo Data Platform
Microsoft read as Microsoft Dynamics 365 Customer Insights
Oracle read as Oracle Unity
Salesforce read as Salesforce Data Cloud
Unresolved, counted raw
Acquia
AudienceStream
Microsoft Customer Data Platform
Discontinued, still offered
No shut-down product was recommended here.