Customer identity and access management
Asked as “customer identity platform”, and as “CIAM solution for a product's users”, on behalf of a mid-market B2B company. 57 first choices recorded across the direct, paraphrase, budget and scale prompts, twelve models each.
Standing · first-choice share
21%
Contested · Auth0 18%
21Frontegg18Auth018Clerk44others
21% of first choices, contested.
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.
The standing
Share is the count of first choices across the direct, paraphrase, budget and scale prompts, over all twelve models, for a mid-market B2B company. Ordered by share.
ProductFirst-choice shareNegative rateLabelsQuadrant
Show the four products at 0%, ordered by negative rate
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.
Firebase Authentication is made by Google, whose model Gemini 3.5 Flash is in the set. On the four share prompts that model made it the first choice zero times of 4; the other eleven models one time of 44. Lab treatment is defined on the method page; the row is marked, not excluded.
Head to head
All twenty-eight head-to-head pages: the top eight products, each against each
Auth0 vs Microsoft Entra External IDAuth0 vs Ping IdentityAuth0 vs FusionAuthAuth0 vs Amazon CognitoWorkOS vs FronteggWorkOS vs ClerkWorkOS vs Microsoft Entra External IDWorkOS vs Ping IdentityWorkOS vs FusionAuthWorkOS vs Amazon CognitoFrontegg vs ClerkFrontegg vs Microsoft Entra External IDFrontegg vs Ping IdentityFrontegg vs FusionAuthFrontegg vs Amazon CognitoClerk vs Microsoft Entra External IDClerk vs Ping IdentityClerk vs FusionAuthClerk vs Amazon CognitoMicrosoft Entra External ID vs Ping IdentityMicrosoft Entra External ID vs FusionAuthMicrosoft Entra External ID vs Amazon CognitoPing Identity vs FusionAuthPing Identity vs Amazon CognitoFusionAuth vs Amazon Cognito
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 →
01
02
03
04
05
06
07
08
09
10
11
12
13
14Accepted challengerEndorsed leader
0%First-choice share → · lines at 30% share and 25% negative40%
Key
01Frontegg21%
Frontegg21%02Auth018%
Auth018%03Clerk18%
Clerk18%04LoginRadius12%
LoginRadius12%05Descope5%
Descope5%06FusionAuth5%
FusionAuth5%07WorkOS4%
WorkOS4%08Microsoft Entra External ID4%
Microsoft Entra External ID4%09Keycloak4%
Keycloak4%10Firebase Authentication2%
Firebase Authentication2%11Stytch0%
Stytch0%12Ping Identity0%
Ping Identity0%13Amazon Cognito0%
Amazon Cognito0%14Okta0%
Okta0%What they warned about
One of twelve models held their first choice under the paraphrase. Claude Haiku 4.5, GPT-5.4 mini, Gemini 3.5 Flash, Perplexity Sonar, Grok 4.20, 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.
Auth022%
13 of 60 labels negative · 9 of 12 models · 1 hard negative
“Budget Trap to Avoid: Auth0 (Okta) ... it is **not recommended for companies on a limited budget**” Gemini 3.5 Flash, budget prompt
Okta50%
10 of 20 labels negative · 9 of 12 models · 1 hard negative
“many models advise small businesses to be cautious or avoid Okta for customer identity use cases due to complexity and cost” Mistral Small, negative prompt
Microsoft Entra External ID23%
9 of 39 labels negative · 7 of 12 models · 5 hard negative
“**Azure AD B2C** if you need a broad, flexible CIAM platform... limited to **single-app external use cases**... **limited customization**” Perplexity Sonar, negative prompt
Amazon Cognito35%
11 of 31 labels negative · 7 of 12 models · 2 hard negative
“Avoid Cognito if your development team wants to move fast with modern, out-of-the-box UI components and smooth SDKs.” Gemini 3.5 Flash, negative prompt
What they cite
Citations exist only for the models that return a source list: fourteen of the twelve in this edition, and all six flagship models on the expanded tier.
Sites the answers cite
78 of 84 answers in this category came back with a source list, from 14 of 12 models: citations where the model returns them, or the search results it consulted. 1175 links across 261 sites, every framing counted. Ranked by the number of answers carrying the site or page.
39 answers · 54 citations · 11 models
36 answers · 61 citations · 11 models
vendor site · WorkOS31 answers · 43 citations · 12 models
30 answers · 30 citations · 11 models
29 answers · 32 citations · 11 models
28 answers · 50 citations · 11 models
vendor site · Guideflow27 answers · 27 citations · 9 models
vendor site · Descope25 answers · 30 citations · 8 models
vendor site · CyberPress23 answers · 32 citations · 10 models
vendor site · LoginRadius22 answers · 39 citations · 11 models
21 answers · 26 citations · 10 models
vendor site · Clerk16 answers · 30 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.
30 answers · 11 models
28 answers · 9 models
27 answers · 9 models
23 answers · 9 models
22 answers · 9 models
20 answers · 8 models
15 answers · 6 models
15 answers · 9 models
14 answers · 9 models
12 answers · 6 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 identity platform”, “customer identity platform”, “customer identity platforms”. 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 |
|---|---|---|---|---|---|---|---|---|
| Frontegg frontegg.com | 21% | 97 | 720 | 4 | 19,060 | 10 | ||
| Auth0 auth0.com Auth0 18%, Auth0/Okta Customer Identity Cloud 0% | 18% | 237 | 18,100 | 772 | 170,673 | 117 | ||
| Clerk clerk.com | 18% | 94 | 110,000 | 59,082 | 46,225 | 12 | ||
| LoginRadius loginradius.com | 12% | 50 | 170 | 7 | 39,709 | 3 | 1 | |
| WorkOS workos.com WorkOS 4%, WorkOS AuthKit 4% | 7% | 99 | 12,100 | 107 | 33,478 | 75 | ||
| Descope descope.com | 5% | 64 | 2,400 | 79 | 19,633 | $192 | 19 | |
| FusionAuth fusionauth.io | 5% | 50 | 1,300 | 21 | 44,199 | 16 | ||
| Microsoft parent site Microsoft Entra External ID 4% | 4% | 93 | 2,240,000 | 403,136 | ||||
| Keycloak keycloak.org | 4% | 54 | 12,100 | 589 | 8,582 | |||
| Supabase Auth supabase.com | 4% | 24 | 1,900 | 62 | 91,424 | 4 |
Google's top ten for the three searches hold 25 sites; 2 of them are among the sites the models cited here (loginradius.com, startwithidentity.com). The first result for “best customer identity platform” is reddit.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 identity platform for a mid-market B2B company?”Paraphrase“Which CIAM solution for a product's users would you recommend to a mid-sized B2B company?”Comparative“What are the top customer identity platforms and how do they differ?”Budget-constrained“What is the best customer identity platform for a company with a limited budget?”Scale-constrained“We are a 500 person company evaluating a customer identity platform. What should we look at?”Negative“Which customer identity platforms should I avoid or be cautious about?”
Claude Haiku 4.5Descope, Fronteggno first choiceChangedno first choiceDescopeno first choiceagainst: Auth0, JumpCloud, LoginRadius, Microsoft Entra, Microsoft Entra ID, Okta, Ping Identity
Three alternatives
OneWelcome Identity Platform, SSOJet, WorkOSTwo alternatives
Auth0, KeycloakGPT-5.4 miniFronteggPingOne for B2BChangedagainst: Microsoft Entra External IDAuth0FusionAuthagainst: Auth0no first choicenothing named
Three alternatives
Auth0/Okta Customer Identity Cloud, FusionAuth, OktaTwo alternatives
Auth0, Microsoft Entra External IDFive alternatives
Amazon Cognito, Clerk, ForgeRock, Ping Identity, PingOneOne alternative
SuperTokensGemini 3.5 FlashWorkOSagainst: Auth0FronteggChangedno first choiceagainst: Auth0, Keycloak, Microsoft Entra External ID, Ping IdentitySupabase Authagainst: Auth0no first choiceagainst: Amazon Cognito, Microsoft Entra External IDagainst: Amazon Cognito, Firebase Authentication, ForgeRock, IBM Verify, Okta/Auth0, Oracle Identity Cloud, Ping Identity
Three alternatives
Clerk, Frontegg, KindeThree alternatives
Auth0, Clerk, WorkOSSix alternatives
Clerk, Descope, Frontegg, Supabase Auth, WorkOS, ZitadelFour alternatives
Clerk, Keycloak, Kinde, ZitadelPerplexity SonarLoginRadiusagainst: Auth0LoginRadius, WorkOSChangedagainst: OktaAuth0Microsoft Entra External IDagainst: Clerkno first choiceagainst: Amazon Cognito, Keycloak, Klaviyo, Microsoft Entra External ID, Simon Data, Tealium, Twilio Segment, mParticle
Three alternatives
Clerk, Frontegg, Microsoft Entra External IDTwo alternatives
Auth0, FronteggNine alternatives
Amazon Cognito, Descope, ForgeRock, FusionAuth, Keycloak, Microsoft Entra External ID, Okta, Ping Identity, StytchOne alternative
FusionAuthGrok 4.20Auth0FronteggChangedno first choiceAuth0, Clerkagainst: Ping IdentityFusionAuth, LoginRadiusagainst: Amazon Cognito, Auth0, ForgeRock, Okta
Six alternatives
Clerk, Frontegg, FusionAuth, Microsoft Entra External ID, Stytch, WorkOSSix alternatives
Amazon Cognito, Auth0, Clerk, LoginRadius, Microsoft Entra External ID, WorkOSSeven alternatives
Amazon Cognito, LoginRadius, Okta/Auth0, Ping Identity, Strivacity, Stytch, Transmit SecurityFour alternatives
Amazon Cognito, FusionAuth, Keycloak, Microsoft Entra External IDOne alternative
Amazon CognitoMistral SmallLoginRadiusagainst: OktaClerk, FronteggChangedno first choiceAuth0no first choiceagainst: Amazon Cognito, ForgeRock, Okta
One alternative
Thales OneWelcomeThree alternatives
LoginRadius, Microsoft Entra External ID, Ping IdentityTwo alternatives
Descope, OktaDeepSeek V4 FlashAuth0, FronteggFronteggChangedagainst: Auth0Auth0Clerk, Keycloakno first choiceagainst: Amazon Cognitoagainst: Amazon Cognito, Firebase Authentication, IBM Verify Identity / CIAM, Microsoft Entra External ID, Okta, Salesforce Identity / Salesforce Headless 360
Three alternatives
Clerk, Microsoft Entra External ID, StytchOne alternative
WorkOSSeven alternatives
Amazon Cognito, Descope, Frontegg, FusionAuth, Keycloak, Microsoft Entra External ID, Ping IdentityThree alternatives
Auth0, FusionAuth, SuperTokensLlama 4 MaverickLoginRadiusno first choiceChangedAuth0no first choiceno first choicenothing named
Two alternatives
Clerk, OneWelcome Identity PlatformOne alternative
Microsoft Entra External IDQwen 3.7 FlashAuth0, Clerkagainst: OktaFronteggChangedagainst: Okta/Auth0no first choiceFirebase Authentication, Supabase Authagainst: Auth0Auth0against: Keycloakagainst: Amazon Cognito, Auth0, IBM Security IDM / Oracle Unified Directory, Microsoft Entra External ID, Microsoft Identity Manager, OneLogin, OpenAM, Ping Identity, Plivo, Stormpath
One alternative
Microsoft Entra IDOne alternative
LoginRadiusSix alternatives
IBM Verify, LoginRadius, Microsoft Entra External ID, Okta, Ping Identity, WorkOSThree alternatives
Clerk, Keycloak, SuperTokensTwo alternatives
Amazon Cognito, Firebase AuthenticationKimi K2Clerk, Fronteggagainst: Amazon Cognito, Auth0FronteggChangedAuth0Auth0, Clerkagainst: Amazon Cognitono first choiceagainst: Auth0, FusionAuthagainst: Amazon Cognito, Firebase Authentication, Okta/Auth0
Two alternatives
FusionAuth, WorkOSThree alternatives
Auth0, Clerk, StytchFour alternatives
Amazon Cognito, LoginRadius, Microsoft Entra External ID, Ping IdentityFour alternatives
Firebase Authentication, FusionAuth, Keycloak, Supabase AuthTwo alternatives
Descope, StytchGLM 4.7 FlashXAuth0Auth0Heldagainst: Ping IdentityAuth0FusionAuth, Keycloakno first choiceagainst: Auth0, ForgeRock, Okta, Ping Identity
Four alternatives
Clerk, Descope, LoginRadius, Microsoft Entra External IDOne alternative
Microsoft Entra External IDThree alternatives
Amazon Cognito, Microsoft Entra External ID, Ping IdentityEight alternatives
Auth0, Clerk, Descope, Firebase Authentication, Hanko, MojoAuth, Supabase Auth, WorkOSMiniMax M2.5Auth0, ClerkClerk, LoginRadiusChangedno first choiceClerk, Descopeno first choiceagainst: IDScan.net, Okta
Two alternatives
Descope, LoginRadiusThree alternatives
Frontegg, Microsoft Entra External ID, PingOne for CustomersSix alternatives
Amazon Cognito, IBM Verify, LoginRadius, Microsoft Entra ID, Okta, Ping IdentityOne alternative
Auth0Bold 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
Seventy-two 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 08:41 | yes | 16 | 10 s |
| Direct recommendation | GPT-5.4 mini | gpt-5.4-mini-2026-03-17 | 2026-10-01 13:36 | yes | 4 | 6 s |
| Direct recommendation | Gemini 3.5 Flash | gemini-3.5-flash | 2026-10-01 10:07 | yes | 27 | 31 s |
| Direct recommendation | Perplexity Sonar | sonar | 2026-10-01 13:49 | yes | 22 | 4 s |
| Direct recommendation | Grok 4.20 | spacexai/grok-4.1-fast-non-reasoning via vertex | 2026-10-01 11:45 | yes | 21 | 16 s |
| Direct recommendation | Mistral Small | mistral/mistral-small via mistral | 2026-10-01 13:47 | yes | 5 | 3 s |
| Direct recommendation | DeepSeek V4 Flash | deepseek/deepseek-v4-flash via deepinfra | 2026-10-01 12:14 | yes | 21 | 47 s |
| Direct recommendation | Llama 4 Maverick | meta/llama-4-maverick via bedrock | 2026-10-01 13:35 | yes | 5 | 2 s |
| Direct recommendation | Qwen 3.7 Flash | alibaba/qwen3.7-flash via alibaba | 2026-10-01 10:03 | no | 0 | 26 s |
| Direct recommendation | Kimi K2 | moonshotai/kimi-k2 via novita | 2026-10-01 12:00 | yes | 14 | 25 s |
| Direct recommendation | GLM 4.7 FlashX | zai/glm-4.7-flashx via zai | 2026-10-01 10:49 | yes | 18 | 65 s |
| Direct recommendation | MiniMax M2.5 | minimax/minimax-m2.5 via minimax | 2026-10-01 10:09 | yes | 10 | 15 s |
| Paraphrase | Claude Haiku 4.5 | claude-haiku-4-5-20251001 | 2026-10-01 08:50 | no | 0 | 4 s |
| Paraphrase | GPT-5.4 mini | gpt-5.4-mini-2026-03-17 | 2026-10-01 08:39 | yes | 4 | 4 s |
| Paraphrase | Gemini 3.5 Flash | gemini-3.5-flash | 2026-10-01 10:15 | yes | 9 | 17 s |
| Paraphrase | Perplexity Sonar | sonar | 2026-10-01 12:19 | yes | 24 | 4 s |
| Paraphrase | Grok 4.20 | spacexai/grok-4.1-fast-non-reasoning via vertex | 2026-10-01 09:15 | yes | 20 | 8 s |
| Paraphrase | Mistral Small | mistral/mistral-small via mistral | 2026-10-01 12:47 | yes | 5 | 4 s |
| Paraphrase | DeepSeek V4 Flash | deepseek/deepseek-v4-flash via deepinfra | 2026-10-01 10:26 | yes | 24 | 27 s |
| Paraphrase | Llama 4 Maverick | meta/llama-4-maverick via bedrock | 2026-10-01 12:46 | yes | 5 | 3 s |
| Paraphrase | Qwen 3.7 Flash | alibaba/qwen3.7-flash via alibaba | 2026-10-01 09:56 | yes | 9 | 15 s |
| Paraphrase | Kimi K2 | moonshotai/kimi-k2 via novita | 2026-10-01 11:41 | yes | 17 | 25 s |
| Paraphrase | GLM 4.7 FlashX | zai/glm-4.7-flashx via zai | 2026-10-01 11:38 | yes | 24 | 113 s |
| Paraphrase | MiniMax M2.5 | minimax/minimax-m2.5 via minimax | 2026-10-01 12:43 | yes | 7 | 39 s |
| Comparative | Claude Haiku 4.5 | claude-haiku-4-5-20251001 | 2026-10-01 09:47 | yes | 15 | 7 s |
| Comparative | GPT-5.4 mini | gpt-5.4-mini-2026-03-17 | 2026-10-01 10:53 | yes | 4 | 8 s |
| Comparative | Gemini 3.5 Flash | gemini-3.5-flash | 2026-10-01 12:42 | yes | 18 | 25 s |
| Comparative | Perplexity Sonar | sonar | 2026-10-01 11:37 | yes | 29 | 7 s |
| Comparative | Grok 4.20 | spacexai/grok-4.1-fast-non-reasoning via vertex | 2026-10-01 10:56 | yes | 22 | 8 s |
| Comparative | Mistral Small | mistral/mistral-small via mistral | 2026-10-01 09:15 | yes | 15 | 12 s |
| Comparative | DeepSeek V4 Flash | deepseek/deepseek-v4-flash via deepinfra | 2026-10-01 08:37 | yes | 21 | 27 s |
| Comparative | Llama 4 Maverick | meta/llama-4-maverick via bedrock | 2026-10-01 13:31 | yes | 5 | 2 s |
| Comparative | Qwen 3.7 Flash | alibaba/qwen3.7-flash via alibaba | 2026-10-01 13:22 | yes | 10 | 23 s |
| Comparative | Kimi K2 | moonshotai/kimi-k2 via novita | 2026-10-01 08:37 | yes | 15 | 24 s |
| Comparative | GLM 4.7 FlashX | zai/glm-4.7-flashx via zai | 2026-10-01 08:34 | yes | 24 | 31 s |
| Comparative | MiniMax M2.5 | minimax/minimax-m2.5 via minimax | 2026-10-01 10:21 | yes | 14 | 24 s |
| Budget constrained | Claude Haiku 4.5 | claude-haiku-4-5-20251001 | 2026-10-01 13:20 | yes | 18 | 8 s |
| Budget constrained | GPT-5.4 mini | gpt-5.4-mini-2026-03-17 | 2026-10-01 13:41 | yes | 3 | 5 s |
| Budget constrained | Gemini 3.5 Flash | gemini-3.5-flash | 2026-10-01 08:34 | yes | 20 | 25 s |
| Budget constrained | Perplexity Sonar | sonar | 2026-10-01 11:10 | yes | 22 | 5 s |
| Budget constrained | Grok 4.20 | spacexai/grok-4.1-fast-non-reasoning via vertex | 2026-10-01 08:29 | yes | 22 | 9 s |
| Budget constrained | Mistral Small | mistral/mistral-small via mistral | 2026-10-01 13:53 | yes | 15 | 5 s |
| Budget constrained | DeepSeek V4 Flash | deepseek/deepseek-v4-flash via deepinfra | 2026-10-01 10:10 | yes | 25 | 22 s |
| Budget constrained | Llama 4 Maverick | meta/llama-4-maverick via bedrock | 2026-10-01 12:00 | yes | 5 | 1 s |
| Budget constrained | Qwen 3.7 Flash | alibaba/qwen3.7-flash via alibaba | 2026-10-01 09:51 | yes | 20 | 51 s |
| Budget constrained | Kimi K2 | moonshotai/kimi-k2 via novita | 2026-10-01 13:09 | yes | 20 | 28 s |
| Budget constrained | GLM 4.7 FlashX | zai/glm-4.7-flashx via zai | 2026-10-01 10:20 | yes | 25 | 126 s |
| Budget constrained | MiniMax M2.5 | minimax/minimax-m2.5 via minimax | 2026-10-01 09:31 | yes | 5 | 13 s |
| Scale constrained | Claude Haiku 4.5 | claude-haiku-4-5-20251001 | 2026-10-01 12:31 | yes | 17 | 10 s |
| Scale constrained | GPT-5.4 mini | gpt-5.4-mini-2026-03-17 | 2026-10-01 08:06 | no | 0 | 9 s |
| Scale constrained | Gemini 3.5 Flash | gemini-3.5-flash | 2026-10-01 07:49 | yes | 11 | 26 s |
| Scale constrained | Perplexity Sonar | sonar | 2026-10-01 12:40 | yes | 30 | 7 s |
| Scale constrained | Grok 4.20 | spacexai/grok-4.1-fast-non-reasoning via vertex | 2026-10-01 07:48 | yes | 19 | 8 s |
| Scale constrained | Mistral Small | mistral/mistral-small via mistral | 2026-10-01 13:41 | no | 0 | 10 s |
| Scale constrained | DeepSeek V4 Flash | deepseek/deepseek-v4-flash via deepinfra | 2026-10-01 08:28 | yes | 24 | 35 s |
| Scale constrained | Llama 4 Maverick | meta/llama-4-maverick via bedrock | 2026-10-01 12:35 | yes | 5 | 1 s |
| Scale constrained | Qwen 3.7 Flash | alibaba/qwen3.7-flash via alibaba | 2026-10-01 08:10 | no | 0 | 30 s |
| Scale constrained | Kimi K2 | moonshotai/kimi-k2 via novita | 2026-10-01 11:06 | yes | 25 | 32 s |
| Scale constrained | GLM 4.7 FlashX | zai/glm-4.7-flashx via zai | 2026-10-01 11:50 | yes | 15 | 86 s |
| Scale constrained | MiniMax M2.5 | minimax/minimax-m2.5 via minimax | 2026-10-01 10:56 | yes | 5 | 25 s |
| Negative framing | Claude Haiku 4.5 | claude-haiku-4-5-20251001 | 2026-10-01 11:46 | yes | 18 | 10 s |
| Negative framing | GPT-5.4 mini | gpt-5.4-mini-2026-03-17 | 2026-10-01 11:21 | yes | 5 | 7 s |
| Negative framing | Gemini 3.5 Flash | gemini-3.5-flash | 2026-10-01 11:30 | yes | 17 | 27 s |
| Negative framing | Perplexity Sonar | sonar | 2026-10-01 11:20 | yes | 23 | 5 s |
| Negative framing | Grok 4.20 | spacexai/grok-4.1-fast-non-reasoning via vertex | 2026-10-01 08:34 | yes | 25 | 7 s |
| Negative framing | Mistral Small | mistral/mistral-small via mistral | 2026-10-01 12:06 | yes | 5 | 3 s |
| Negative framing | DeepSeek V4 Flash | deepseek/deepseek-v4-flash via deepinfra | 2026-10-01 08:04 | yes | 22 | 28 s |
| Negative framing | Llama 4 Maverick | meta/llama-4-maverick via bedrock | 2026-10-01 08:18 | yes | 5 | 1 s |
| Negative framing | Qwen 3.7 Flash | alibaba/qwen3.7-flash via alibaba | 2026-10-01 13:39 | no | 0 | 40 s |
| Negative framing | Kimi K2 | moonshotai/kimi-k2 via novita | 2026-10-01 08:13 | yes | 25 | 24 s |
| Negative framing | GLM 4.7 FlashX | zai/glm-4.7-flashx via zai | 2026-10-01 12:58 | yes | 25 | 36 s |
| Negative framing | MiniMax M2.5 | minimax/minimax-m2.5 via minimax | 2026-10-01 11:34 | yes | 15 | 26 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
3 of 4 flipped
GPT-5.4 mini
3 of 4 flipped
Gemini 3.5 Flash
2 of 4 flipped
Perplexity Sonar
4 of 4 flipped
Grok 4.20
5 of 6 flipped
Mistral Small
2 of 3 flipped
DeepSeek V4 Flash
3 of 4 flipped
Llama 4 Maverick
2 of 3 flipped
Qwen 3.7 Flash
6 of 6 flipped
Kimi K2
4 of 5 flipped
GLM 4.7 FlashX
4 of 4 flipped
MiniMax M2.5
4 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
Ping read as Ping Identity
Unresolved, counted raw
IBM Security IDM / Oracle Unified Directory
IBM Verify Identity / CIAM
Klaviyo
OpenAM (ForgeRock)
Oracle Identity Cloud
Passage by 1Password
PingOne for B2B
PocketBase/Directus
Salesforce Identity / Salesforce Headless 360
Splink
Stormpath
Truss ID
Zingg
Discontinued, still offered
No shut-down product was recommended here.