Compensation benchmarking data
Asked as “salary benchmarking data provider”, and as “compensation survey data”, on behalf of a mid-market B2B company. 58 first choices recorded across the direct, paraphrase, budget and scale prompts, fourteen models each.
Standing · first-choice share
19Pave19Payscale10SalaryCube52others
19% of first choices, contested.
Since September 2026=heldSince September 2026: 20% → 20%, ±0 points. Inside the 13-point floor: within noise. Read over the models both editions asked.Pave held the lead, ±0 points on 20%, inside the 13-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
Pave30%Contested▼−2Since September 2026: 32% → 30%, −2 points. Inside the 13-point floor: within noise. Read over the models both editions asked.Standing →Mid-marketPave19%Contested▼−2Since September 2026: 24% → 22%, −2 points. Inside the 13-point floor: within noise. Read over the models both editions asked.This pageEnterprise
Mercer Total Remuneration Survey44%Clear leader▲+14Since September 2026: 30% → 44%, +14 points. Past the 13-point floor: movement. Read over the models both editions asked.Standing →
Pave30%Contested▼−2Since September 2026: 32% → 30%, −2 points. Inside the 13-point floor: within noise. Read over the models both editions asked.Standing →Mid-marketPave19%Contested▼−2Since September 2026: 24% → 22%, −2 points. Inside the 13-point floor: within noise. Read over the models both editions asked.This pageEnterpriseMercer Total Remuneration Survey44%Clear leader▲+14Since September 2026: 30% → 44%, +14 points. Past the 13-point floor: movement. 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
01Pave19%4%27accepted challenger=heldSince September 2026: 20% → 20%, ±0 points. Inside the 13-point floor: within noise. Read over the models both editions asked.20% → 20%
Pave19%4%27accepted challenger=heldSince September 2026: 20% → 20%, ±0 points. Inside the 13-point floor: within noise. Read over the models both editions asked.20% → 20%02
Payscale19%14%35accepted challenger▼−2Since September 2026: 24% → 22%, −2 points. Inside the 13-point floor: within noise. Read over the models both editions asked.24% → 22%
Payscale19%14%35accepted challenger▼−2Since September 2026: 24% → 22%, −2 points. Inside the 13-point floor: within noise. Read over the models both editions asked.24% → 22%03
SalaryCube10%8%12accepted challenger▲+6Since September 2026: 4% → 10%, +6 points. Inside the 13-point floor: within noise. Read over the models both editions asked.4% → 10%
SalaryCube10%8%12accepted challenger▲+6Since September 2026: 4% → 10%, +6 points. Inside the 13-point floor: within noise. Read over the models both editions asked.4% → 10%04
Salary.com CompAnalyst9%4%24accepted challenger▲+4Since September 2026: 2% → 6%, +4 points. Inside the 13-point floor: within noise. Read over the models both editions asked.2% → 6%
Salary.com CompAnalyst9%4%24accepted challenger▲+4Since September 2026: 2% → 6%, +4 points. Inside the 13-point floor: within noise. Read over the models both editions asked.2% → 6%05
Ravio7%4%26accepted challenger▲+8Since September 2026: 0% → 8%, +8 points. Inside the 13-point floor: within noise. Read over the models both editions asked.0% → 8%
Ravio7%4%26accepted challenger▲+8Since September 2026: 0% → 8%, +8 points. Inside the 13-point floor: within noise. Read over the models both editions asked.0% → 8%06
Comprehensive.io7%0%10accepted challenger▼−4Since September 2026: 12% → 8%, −4 points. Inside the 13-point floor: within noise. Read over the models both editions asked.12% → 8%
Comprehensive.io7%0%10accepted challenger▼−4Since September 2026: 12% → 8%, −4 points. Inside the 13-point floor: within noise. Read over the models both editions asked.12% → 8%07Radford5%50%28criticized challenger▲+2Since September 2026: 4% → 6%, +2 points. Inside the 13-point floor: within noise. Read over the models both editions asked.4% → 6%
08Mercer Total Remuneration Survey3%52%33criticized challenger▼−4Since September 2026: 6% → 2%, −4 points. Inside the 13-point floor: within noise. Read over the models both editions asked.6% → 2%
Mercer Total Remuneration Survey3%52%33criticized challenger▼−4Since September 2026: 6% → 2%, −4 points. Inside the 13-point floor: within noise. Read over the models both editions asked.6% → 2%09
WTW2%58%12criticized challenger=heldSince September 2026: 0% → 0%, ±0 points. Inside the 13-point floor: within noise. Read over the models both editions asked.0% → 0%
WTW2%58%12criticized challenger=heldSince September 2026: 0% → 0%, ±0 points. Inside the 13-point floor: within noise. Read over the models both editions asked.0% → 0%Show the four products at 0%, ordered by negative rate
13
Glassdoor0%100%10criticized challenger=heldSince September 2026: 0% → 0%, ±0 points. Inside the 13-point floor: within noise. Read over the models both editions asked.0% → 0%
Glassdoor0%100%10criticized challenger=heldSince September 2026: 0% → 0%, ±0 points. Inside the 13-point floor: within noise. Read over the models both editions asked.0% → 0%10
Figures0%0%13accepted challenger=heldSince September 2026: 0% → 0%, ±0 points. Inside the 13-point floor: within noise. Read over the models both editions asked.0% → 0%
Figures0%0%13accepted challenger=heldSince September 2026: 0% → 0%, ±0 points. Inside the 13-point floor: within noise. Read over the models both editions asked.0% → 0%11
Gartner0%0%12accepted challenger=heldSince September 2026: 0% → 0%, ±0 points. Inside the 13-point floor: within noise. Read over the models both editions asked.0% → 0%
Gartner0%0%12accepted challenger=heldSince September 2026: 0% → 0%, ±0 points. Inside the 13-point floor: within noise. Read over the models both editions asked.0% → 0%12
APQC0%0%10accepted challenger=heldSince September 2026: 0% → 0%, ±0 points. Inside the 13-point floor: within noise. Read over the models both editions asked.0% → 0%
APQC0%0%10accepted challenger=heldSince September 2026: 0% → 0%, ±0 points. Inside the 13-point floor: within noise. Read over the models both editions asked.0% → 0%The floor is 13 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-eight head-to-head pages: the top eight products, each against each
Pave vs RavioPave vs Comprehensive.ioPave vs RadfordPave vs Mercer Total Remuneration SurveyPayscale vs SalaryCubePayscale vs Salary.com CompAnalystPayscale vs RavioPayscale vs Comprehensive.ioPayscale vs RadfordPayscale vs Mercer Total Remuneration SurveySalaryCube vs Salary.com CompAnalystSalaryCube vs RavioSalaryCube vs Comprehensive.ioSalaryCube vs RadfordSalaryCube vs Mercer Total Remuneration SurveySalary.com CompAnalyst vs RavioSalary.com CompAnalyst vs Comprehensive.ioSalary.com CompAnalyst vs RadfordSalary.com CompAnalyst vs Mercer Total Remuneration SurveyRavio vs Comprehensive.ioRavio vs RadfordRavio vs Mercer Total Remuneration SurveyComprehensive.io vs RadfordComprehensive.io vs Mercer Total Remuneration SurveyRadford vs Mercer Total Remuneration Survey
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 →
010203040506R07
080910111213Accepted challengerEndorsed leader
0%First-choice share → · lines at 30% share and 25% negative30%
Key
01Pave19%
Pave19%02Payscale19%
Payscale19%03SalaryCube10%
SalaryCube10%04Salary.com CompAnalyst9%
Salary.com CompAnalyst9%05Ravio7%
Ravio7%06Comprehensive.io7%
Comprehensive.io7%07Radford5%
08Mercer Total Remuneration Survey3%
Mercer Total Remuneration Survey3%09WTW2%
WTW2%10Figures0%
Figures0%11Gartner0%
Gartner0%12APQC0%
APQC0%13Glassdoor0%
Glassdoor0%What they warned about
Three of fourteen models held their first choice under the paraphrase. GPT-5.4 mini, Gemini 3.5 Flash, Perplexity Sonar, Grok 4.1 Fast, Mistral Small, Llama 4 Maverick, Qwen 3.7 Flash, Kimi K2, GLM 4.7 FlashX, GPT-6 Luna and Muse Glimmer 30B 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.
Mercer Total Remuneration Survey52%
17 of 33 labels negative · 9 of 14 models · 4 hard negative
“**Avoid:** Traditional survey providers like Mercer, Korn Ferry, or Willis Towers Watson” GLM 4.7 FlashX, paraphrase prompt
Radford
50%
14 of 28 labels negative · 9 of 14 models · 1 hard negative
“"Why you should avoid traditional providers (Mercer, Radford, Willis Towers Watson)"” Muse Glimmer 30B, negative prompt
Glassdoor100%
10 of 10 labels negative · 6 of 14 models · 3 hard negative
“Avoid providers that rely heavily on employee self-reporting (like Glassdoor data) or scraped job postings, which are often inflated or inaccurate.” Qwen 3.7 Flash, scale prompt
Willis Towers Watson
56%
5 of 9 labels negative · 5 of 14 models · 4 hard negative
“**Avoid:** Traditional survey providers like Mercer, Korn Ferry, or Willis Towers Watson” GLM 4.7 FlashX, paraphrase 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
74 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. 1009 links across 241 sites, every framing counted. Ranked by the number of answers carrying the site or page. 28 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.
vendor site · Ravio49 answers · 134 citations · 12 models
28 answers · 66 citations · 11 models
vendor site · Comprehensive28 answers · 35 citations · 11 models
28 answers · 28 citations · 12 models
vendor site · Pin24 answers · 24 citations · 12 models
vendor site · SalaryCube23 answers · 49 citations · 11 models
vendor site · Deel23 answers · 25 citations · 10 models
22 answers · 42 citations · 11 models
16 answers · 36 citations · 10 models
vendor site · Figures16 answers · 19 citations · 9 models
vendor site · G215 answers · 20 citations · 8 models
vendor site · CaptivateIQ14 answers · 15 citations · 10 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.
35 answers · 11 models
28 answers · 12 models
24 answers · 12 models
23 answers · 11 models
19 answers · 10 models
16 answers · 9 models
15 answers · 7 models
14 answers · 10 models
13 answers · 8 models
12 answers · 9 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 “benchmarking providers”, “best salary benchmarking data provider”, “salary benchmarking data provider”. 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 |
|---|---|---|---|---|---|---|---|---|
| Pave pave.com Pave 19%, Pave Market Data Lite 2% | 21% | 213 | 27,100 | 28,423 | 7,645 | $91 | 7 | |
| Payscale payscale.com Payscale 19%, Payscale Ascent 2% | 21% | 229 | 9,900 | 299 | 1,408,989 | $26 | 33 | 3 |
| SalaryCube salarycube.com | 10% | 37 | 30 | 0 | 2,822 | 1 | 8 | |
| Salary.com CompAnalyst salary.com | 9% | 90 | 260 | 3 | 1,069,118 | $1,642 | 70 | |
| Ravio ravio.com | 7% | 165 | 1,900 | 181 | 8,419 | 1 | ||
| Comprehensive.io comprehensive.io | 7% | 55 | 50 | 0 | 2,021 | 3 | ||
| Radford | 5% | 18,100 | 5,850 | |||||
| Mercer mercer.com Mercer Total Remuneration Survey 3%, Mercer Benchmark Database 2% | 5% | 106 | 49,500 | 23,681 | 98,058 company | 18 | 4 | |
| CompData Surveys compdata.ai | 3% | 3 | ||||||
| Lattice lattice.com | 2% | 662 | 90,500 | 13,900 | 122,123 | $1,328 | 120 |
Google's top ten for the three searches hold 25 sites; 5 of them are among the sites the models cited here (captivateiq.com, comprehensive.io, g2.com, ravio.com, salarycube.com). The first result for “benchmarking providers” is thehackettgroup.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 salary benchmarking data provider for a mid-market B2B company?”Paraphrase“Which compensation survey data would you recommend to a mid-sized B2B company?”Comparative“What are the top benchmarking providers and how do they differ?”Budget-constrained“What is the best salary benchmarking data provider for a company with a limited budget?”Scale-constrained“We are a 500 person company evaluating a salary benchmarking data provider. What should we look at?”Negative“Which benchmarking providers should I avoid or be cautious about?”
Claude Haiku 4.5PayscalePayscaleHeldagainst: Mercer Total Remuneration Surveyno first choiceComprehensive.io, Paveno first choiceagainst: UserBenchmark
Four alternatives
Comprehensive, Lattice, Pave, SalaryCubeThree alternatives
Ravio, Salary.com CompAnalyst, Willis Towers WatsonThree alternatives
Bureau of Labor Statistics, Payscale, Salary.com CompAnalystGPT-5.4 miniPayscaleno first choiceChangedno first choiceMercer MarketPricer, Payscaleagainst: Culpepper Compensation Surveys, Korn Ferry Pay, Mercer Total Remuneration Survey, Radford, WTWno first choicenothing named
Three alternatives
Mercer Total Remuneration Survey, Radford, WTWGemini 3.5 FlashPaveagainst: Aon Radford, Mercer Total Remuneration SurveyPave, PayscaleChangedagainst: RadfordMercer Total Remuneration Survey, Willis Towers WatsonPaveagainst: Glassdoor, Mercer Total Remuneration Survey, Radford, WTWno first choiceagainst: Glassdoor, Levels.fyi, Payscaleagainst: Glassdoor, Korn Ferry Pay, MGMA, Payscale, Radford, Sullivan Cotter
Five alternatives
Figures, Payscale, Ravio, RepVue, Salary.com CompAnalystFour alternatives
Carta Total Compensation, Figures, Ravio, Salary.com CompAnalystFive alternatives
Bureau of Labor Statistics (BLS) - OEWS, Comprehensive.io, Figures, OpenComp, RavioPerplexity SonarRavioCompData Surveys, Xactly's/Benchmarkit's B2B sales compensation surveyChangedagainst: PayscaleMercer Total Remuneration SurveyPaveno first choiceagainst: Korn Ferry Pay, Radford, Willis Towers Watson
Five alternatives
CompUp, Pave, Payfactors, Payscale Ascent, Salary.com CompAnalystTwo alternatives
BDO private company executive compensation survey, Mercer SME/market surveysFour alternatives
BLS OEWS, Comprehensive.io, Payscale, RavioGrok 4.1 FastSalaryCubeagainst: Korn Ferry Pay, Lattice, Mercer Total Remuneration Survey, Ravio, WTWSalary.com CompAnalystChangedagainst: Radford, WTWno first choiceComprehensive.ioagainst: Glassdoorno first choiceagainst: Glassdooragainst: GSM8K, HumanEval, LMSYS Chatbot Arena, MLPerf, MMLU, SPEC, SWE-bench, TPC, Terminal-Bench, WebArena
Two alternatives
Payscale, Salary.com CompAnalystTwo alternatives
Culpepper Compensation Surveys, PayscaleFive alternatives
Pave, Payscale, Ravio, Salary.com CompAnalyst, U.S. Bureau of Labor Statistics (BLS) Occupational Employment StatisticsMistral SmallRavio, SalaryCubeMercer Benchmark Database, RadfordChangedThe Conference BoardPayscale, Radfordno first choiceagainst: UserBenchmark
Two alternatives
CompUp, FiguresTwo alternatives
Bureau of Labor Statistics (BLS) – Occupational Employment and Wage Statistics, PayscaleSix alternatives
APQC, Computer Economics, Deloitte, Gartner, McKinsey & Company, OxeraDeepSeek V4 FlashPayscaleagainst: Mercer Total Remuneration Survey, Radford, SalaryCubePayscaleHeldagainst: Mercer Total Remuneration Survey, Radfordno first choiceComprehensive.io, Paveagainst: Aon's Radford McLagan Compensation Database, Mercer Total Remuneration Survey, Willis Towers Watsonno first choiceagainst: Paveagainst: Aon's Radford McLagan Compensation Database, Brightmine, Mercer Total Remuneration Survey, Radford, WTW
Two alternatives
Pave, Salary.com CompAnalystTwo alternatives
Betts Recruiting Compensation Guide, Pavilion's B2B GTM Compensation BenchmarksSeven alternatives
Avasant, Computer Economics, Deloitte, Forrester, Gartner, ISG, KPMGTwo alternatives
BLS Occupational Employment & Wage Statistics, RavioLlama 4 Maverickno first choiceMercer Total Remuneration Survey, PayscaleChangedno first choicePaveno first choiceagainst: Glassdoor, Hays
Two alternatives
ravio.com, salarycube.comTwo alternatives
Deloitte, McKinsey & CompanyTwo alternatives
BLS OEWS, Comprehensive.ioQwen 3.7 FlashLatticeagainst: Mercer Total Remuneration Survey, RadfordPayscaleChangedagainst: Aon's Radford McLagan Compensation Database, Mercer Total Remuneration Survey, WTWMercer Total Remuneration SurveySalary.com CompAnalystagainst: Mercer Total Remuneration Survey, Willis Towers Watsonno first choiceagainst: Glassdoor, Gusto, Mercer Total Remuneration Survey, Radford, WTW, Workdaynothing named
Four alternatives
Comprehensive, Deel HR, Pave, SalaryCubeThree alternatives
Korn Ferry Pay, Radford, Robert HalfTwo alternatives
Aon's Radford McLagan Compensation Database, RadfordThree alternatives
Pave, Ravio, U.S. Bureau of Labor Statistics (BLS) Occupational Employment and Wage StatisticsThree alternatives
Lattice, Payscale, RavioKimi K2Payscale, SalaryCubeagainst: RadfordRadford, Salary.com CompAnalystChangedno first choicePave, Ravio, U.S. Bureau of Labor Statisticsagainst: Glassdoorno first choiceagainst: GAIA, LINPACK, LMSYS Chatbot Arena, OSWorld, SWE-bench, Terminal-Bench, WebArena, Whetstone
Four alternatives
Figures, Pave, Ravio, Salary.com CompAnalystSeven alternatives
Bridge Group, Mercer Total Remuneration Survey, Pave, Payfactors, Payscale, SalaryCube, Workleap CompensationFour alternatives
Deel Compensation, Option Impact, Pequity, SalaryExpertFive alternatives
Figures, Lattice, Pave, Payscale, Salary.com CompAnalystGLM 4.7 FlashXPave, Payscale Ascentagainst: Mercer Total Remuneration SurveyRavioChangedagainst: Glassdoor, Indeed, Korn Ferry Pay, Mercer Total Remuneration Survey, Willis Towers Watsonno first choiceBLS, Comprehensive.io, Paveagainst: Glassdoor, Indeedno first choicenothing named
Two alternatives
Figures, Salary.com CompAnalystThree alternatives
Bureau of Labor Statistics, Lattice, SalaryCubeSeven alternatives
APQC, Accenture, Deloitte, Forrester, Gartner, Grant Thornton, McKinsey & CompanyFive alternatives
Figures, Levels.fyi, Payscale, Ravio, Salary.com CompAnalystMiniMax M2.5SalaryCubeSalaryCubeHeldno first choiceBureau of Labor Statisticsagainst: Payscale, Salary.com CompAnalystno first choiceagainst: Mercer Total Remuneration Survey, Radfordnothing named
Three alternatives
CompUp, Lattice, Salary.com CompAnalystFive alternatives
Mercer Total Remuneration Survey, Pave, Payscale, Radford McLagan, RavioThree alternatives
Comprehensive.io, Indeed, PaveFour alternatives
CompAnalyst, Figures, Payscale, RavioGPT-6 LunaPaveMercer Total Remuneration Survey, WTWChangedAon's Radford McLagan Compensation Database, Mercer Total Remuneration Survey, WTWSalary.com CompAnalystno first choicenothing named
Three alternatives
Mercer Total Remuneration Survey, Payscale, RadfordTwo alternatives
Aon Radford, PayscaleFour alternatives
APQC, Gartner, ISG, The Hackett GroupOne alternative
Bureau of Labor Statistics' free wage dataMuse Glimmer 30BSalaryCubeCompData Surveys, Salary.com CompAnalystChangedDeloittePave Market Data Liteno first choiceagainst: Mercer Total Remuneration Survey, Payscale, Radford, UserBenchmark, Willis Towers Watson
Two alternatives
Pave, Salary.com CompAnalystFour alternatives
Mercer Total Remuneration Survey TRS / Management Benchmarking Database / Comptryx, Payscale Knowledge / Payscale Salary Survey, Radford, Aon, SalaryCubeThree alternatives
APQC, Computer Economics, GartnerTwo alternatives
BLS Occupational Employment and Wage Statistics OEWS, Comprehensive.ioBold 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 08:26 | yes | 9 | 7 s |
| Direct recommendation | GPT-5.4 mini | gpt-5.4-mini-2026-03-17 | 2026-10-01 11:17 | yes | 3 | 6 s |
| Direct recommendation | Gemini 3.5 Flash | gemini-3.5-flash | 2026-10-01 10:51 | yes | 16 | 26 s |
| Direct recommendation | Perplexity Sonar | sonar | 2026-10-01 10:08 | yes | 20 | 4 s |
| Direct recommendation | Grok 4.1 Fast | spacexai/grok-4.1-fast-non-reasoning via vertex | 2026-10-01 10:51 | yes | 19 | 8 s |
| Direct recommendation | Mistral Small | mistral/mistral-small via mistral | 2026-10-01 11:14 | yes | 11 | 5 s |
| Direct recommendation | DeepSeek V4 Flash | deepseek/deepseek-v4-flash via deepinfra | 2026-10-01 10:09 | yes | 18 | 26 s |
| Direct recommendation | Llama 4 Maverick | meta/llama-4-maverick via bedrock | 2026-10-01 08:08 | yes | 5 | 2 s |
| Direct recommendation | Qwen 3.7 Flash | alibaba/qwen3.7-flash via alibaba | 2026-10-01 10:01 | yes | 8 | 55 s |
| Direct recommendation | Kimi K2 | moonshotai/kimi-k2 via novita | 2026-10-01 10:06 | yes | 14 | 20 s |
| Direct recommendation | GLM 4.7 FlashX | zai/glm-4.7-flashx via zai | 2026-10-01 09:36 | yes | 12 | 310 s |
| Direct recommendation | MiniMax M2.5 | minimax/minimax-m2.5 via minimax | 2026-10-01 09:27 | yes | 5 | 39 s |
| Direct recommendation | GPT-6 Luna | gpt-6-luna | 2026-10-01 07:29 | yes | 3 | 18 s |
| Direct recommendation | Muse Glimmer 30B | meta/muse-glimmer-30b via togetherai | 2026-10-01 07:51 | yes | 22 | 53 s |
| Paraphrase | Claude Haiku 4.5 | claude-haiku-4-5-20251001 | 2026-10-01 07:53 | yes | 17 | 10 s |
| Paraphrase | GPT-5.4 mini | gpt-5.4-mini-2026-03-17 | 2026-10-01 12:08 | no | 0 | 6 s |
| Paraphrase | Gemini 3.5 Flash | gemini-3.5-flash | 2026-10-01 09:11 | yes | 12 | 22 s |
| Paraphrase | Perplexity Sonar | sonar | 2026-10-01 09:17 | yes | 20 | 3 s |
| Paraphrase | Grok 4.1 Fast | spacexai/grok-4.1-fast-non-reasoning via vertex | 2026-10-01 09:06 | yes | 18 | 13 s |
| Paraphrase | Mistral Small | mistral/mistral-small via mistral | 2026-10-01 11:11 | no | 0 | 6 s |
| Paraphrase | DeepSeek V4 Flash | deepseek/deepseek-v4-flash via deepinfra | 2026-10-01 07:45 | yes | 22 | 28 s |
| Paraphrase | Llama 4 Maverick | meta/llama-4-maverick via bedrock | 2026-10-01 07:42 | yes | 5 | 1 s |
| Paraphrase | Qwen 3.7 Flash | alibaba/qwen3.7-flash via alibaba | 2026-10-01 11:53 | no | 0 | 37 s |
| Paraphrase | Kimi K2 | moonshotai/kimi-k2 via novita | 2026-10-01 08:44 | yes | 13 | 26 s |
| Paraphrase | GLM 4.7 FlashX | zai/glm-4.7-flashx via zai | 2026-10-01 09:34 | yes | 21 | 39 s |
| Paraphrase | MiniMax M2.5 | minimax/minimax-m2.5 via minimax | 2026-10-01 11:17 | yes | 7 | 25 s |
| Paraphrase | GPT-6 Luna | gpt-6-luna | 2026-10-01 08:01 | yes | 5 | 33 s |
| Paraphrase | Muse Glimmer 30B | meta/muse-glimmer-30b via togetherai | 2026-10-01 09:57 | yes | 12 | 34 s |
| Comparative | Claude Haiku 4.5 | claude-haiku-4-5-20251001 | 2026-10-01 09:18 | yes | 17 | 9 s |
| Comparative | GPT-5.4 mini | gpt-5.4-mini-2026-03-17 | 2026-10-01 09:07 | no | 0 | 6 s |
| Comparative | Gemini 3.5 Flash | gemini-3.5-flash | 2026-10-01 09:47 | yes | 18 | 23 s |
| Comparative | Perplexity Sonar | sonar | 2026-10-01 09:15 | yes | 21 | 6 s |
| Comparative | Grok 4.1 Fast | spacexai/grok-4.1-fast-non-reasoning via vertex | 2026-10-01 08:49 | yes | 21 | 11 s |
| Comparative | Mistral Small | mistral/mistral-small via mistral | 2026-10-01 11:05 | yes | 5 | 6 s |
| Comparative | DeepSeek V4 Flash | deepseek/deepseek-v4-flash via deepinfra | 2026-10-01 07:53 | yes | 24 | 42 s |
| Comparative | Llama 4 Maverick | meta/llama-4-maverick via bedrock | 2026-10-01 11:13 | yes | 5 | 10 s |
| Comparative | Qwen 3.7 Flash | alibaba/qwen3.7-flash via alibaba | 2026-10-01 09:08 | yes | 16 | 32 s |
| Comparative | Kimi K2 | moonshotai/kimi-k2 via novita | 2026-10-01 11:53 | yes | 25 | 36 s |
| Comparative | GLM 4.7 FlashX | zai/glm-4.7-flashx via zai | 2026-10-01 11:48 | yes | 23 | 32 s |
| Comparative | MiniMax M2.5 | minimax/minimax-m2.5 via minimax | 2026-10-01 07:56 | yes | 5 | 50 s |
| Comparative | GPT-6 Luna | gpt-6-luna | 2026-10-01 09:41 | yes | 6 | 13 s |
| Comparative | Muse Glimmer 30B | meta/muse-glimmer-30b via togetherai | 2026-10-01 08:18 | yes | 14 | 31 s |
| Budget constrained | Claude Haiku 4.5 | claude-haiku-4-5-20251001 | 2026-10-01 08:00 | yes | 16 | 10 s |
| Budget constrained | GPT-5.4 mini | gpt-5.4-mini-2026-03-17 | 2026-10-01 08:14 | yes | 3 | 5 s |
| Budget constrained | Gemini 3.5 Flash | gemini-3.5-flash | 2026-10-01 08:24 | yes | 19 | 19 s |
| Budget constrained | Perplexity Sonar | sonar | 2026-10-01 10:05 | yes | 20 | 3 s |
| Budget constrained | Grok 4.1 Fast | spacexai/grok-4.1-fast-non-reasoning via vertex | 2026-10-01 09:10 | yes | 19 | 12 s |
| Budget constrained | Mistral Small | mistral/mistral-small via mistral | 2026-10-01 08:16 | yes | 5 | 3 s |
| Budget constrained | DeepSeek V4 Flash | deepseek/deepseek-v4-flash via deepinfra | 2026-10-01 12:01 | yes | 24 | 31 s |
| Budget constrained | Llama 4 Maverick | meta/llama-4-maverick via bedrock | 2026-10-01 12:09 | yes | 5 | 2 s |
| Budget constrained | Qwen 3.7 Flash | alibaba/qwen3.7-flash via alibaba | 2026-10-01 09:25 | yes | 5 | 24 s |
| Budget constrained | Kimi K2 | moonshotai/kimi-k2 via novita | 2026-10-01 10:07 | yes | 9 | 19 s |
| Budget constrained | GLM 4.7 FlashX | zai/glm-4.7-flashx via zai | 2026-10-01 09:01 | yes | 23 | 29 s |
| Budget constrained | MiniMax M2.5 | minimax/minimax-m2.5 via minimax | 2026-10-01 11:16 | yes | 8 | 32 s |
| Budget constrained | GPT-6 Luna | gpt-6-luna | 2026-10-01 08:57 | yes | 3 | 14 s |
| Budget constrained | Muse Glimmer 30B | meta/muse-glimmer-30b via togetherai | 2026-10-01 09:11 | yes | 22 | 27 s |
| Scale constrained | Claude Haiku 4.5 | claude-haiku-4-5-20251001 | 2026-10-01 08:55 | no | 0 | 6 s |
| Scale constrained | GPT-5.4 mini | gpt-5.4-mini-2026-03-17 | 2026-10-01 09:50 | no | 0 | 9 s |
| Scale constrained | Gemini 3.5 Flash | gemini-3.5-flash | 2026-10-01 12:17 | no | 0 | 51 s |
| Scale constrained | Perplexity Sonar | sonar | 2026-10-01 08:04 | yes | 20 | 6 s |
| Scale constrained | Grok 4.1 Fast | spacexai/grok-4.1-fast-non-reasoning via vertex | 2026-10-01 09:35 | yes | 13 | 7 s |
| Scale constrained | Mistral Small | mistral/mistral-small via mistral | 2026-10-01 08:28 | no | 0 | 6 s |
| Scale constrained | DeepSeek V4 Flash | deepseek/deepseek-v4-flash via deepinfra | 2026-10-01 11:50 | yes | 16 | 39 s |
| Scale constrained | Llama 4 Maverick | meta/llama-4-maverick via bedrock | 2026-10-01 08:07 | yes | 5 | 2 s |
| Scale constrained | Qwen 3.7 Flash | alibaba/qwen3.7-flash via alibaba | 2026-10-01 11:57 | yes | 8 | 47 s |
| Scale constrained | Kimi K2 | moonshotai/kimi-k2 via novita | 2026-10-01 10:54 | yes | 19 | 25 s |
| Scale constrained | GLM 4.7 FlashX | zai/glm-4.7-flashx via zai | 2026-10-01 09:10 | yes | 18 | 93 s |
| Scale constrained | MiniMax M2.5 | minimax/minimax-m2.5 via minimax | 2026-10-01 07:46 | yes | 5 | 33 s |
| Scale constrained | GPT-6 Luna | gpt-6-luna | 2026-10-01 08:09 | yes | 3 | 19 s |
| Scale constrained | Muse Glimmer 30B | meta/muse-glimmer-30b via togetherai | 2026-10-01 11:55 | yes | 12 | 21 s |
| Negative framing | Claude Haiku 4.5 | claude-haiku-4-5-20251001 | 2026-10-01 09:40 | yes | 20 | 8 s |
| Negative framing | GPT-5.4 mini | gpt-5.4-mini-2026-03-17 | 2026-10-01 09:40 | yes | 5 | 7 s |
| Negative framing | Gemini 3.5 Flash | gemini-3.5-flash | 2026-10-01 10:55 | yes | 12 | 22 s |
| Negative framing | Perplexity Sonar | sonar | 2026-10-01 07:33 | yes | 39 | 4 s |
| Negative framing | Grok 4.1 Fast | spacexai/grok-4.1-fast-non-reasoning via vertex | 2026-10-01 08:23 | yes | 24 | 9 s |
| Negative framing | Mistral Small | mistral/mistral-small via mistral | 2026-10-01 09:20 | yes | 5 | 5 s |
| Negative framing | DeepSeek V4 Flash | deepseek/deepseek-v4-flash via deepinfra | 2026-10-01 10:57 | yes | 21 | 33 s |
| Negative framing | Llama 4 Maverick | meta/llama-4-maverick via bedrock | 2026-10-01 08:41 | yes | 5 | 2 s |
| Negative framing | Qwen 3.7 Flash | alibaba/qwen3.7-flash via alibaba | 2026-10-01 11:04 | yes | 10 | 30 s |
| Negative framing | Kimi K2 | moonshotai/kimi-k2 via novita | 2026-10-01 11:29 | yes | 25 | 27 s |
| Negative framing | GLM 4.7 FlashX | zai/glm-4.7-flashx via zai | 2026-10-01 11:44 | no | 0 | 24 s |
| Negative framing | MiniMax M2.5 | minimax/minimax-m2.5 via minimax | 2026-10-01 08:57 | yes | 13 | 28 s |
| Negative framing | GPT-6 Luna | gpt-6-luna | 2026-10-01 08:29 | no | 0 | 3 s |
| Negative framing | Muse Glimmer 30B | meta/muse-glimmer-30b via togetherai | 2026-10-01 07:39 | yes | 13 | 20 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 3 flipped
GPT-5.4 mini
4 of 4 flipped
Gemini 3.5 Flash
2 of 4 flipped
Perplexity Sonar
3 of 5 flipped
Grok 4.1 Fast
2 of 3 flipped
Mistral Small
3 of 4 flipped
DeepSeek V4 Flash
2 of 4 flipped
Llama 4 Maverick
1 of 2 flipped
Qwen 3.7 Flash
4 of 4 flipped
Kimi K2
3 of 4 flipped
GLM 4.7 FlashX
4 of 4 flipped
MiniMax M2.5
3 of 3 flipped
GPT-6 Luna
4 of 4 flipped
Muse Glimmer 30B
3 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
Aon read as Aon's Radford McLagan Compensation Database
Bain read as Bain & Company
Carta Total Comp read as Carta Total Compensation
Culpepper read as Culpepper Compensation Surveys
Korn Ferry read as Korn Ferry Pay
Korn Ferry (Hay Group) read as Korn Ferry Pay
McKinsey read as McKinsey & Company
Mercer read as Mercer Total Remuneration Survey
Salary.com read as Salary.com CompAnalyst
Salary.com (CompAnalyst) read as Salary.com CompAnalyst
Salary.com (CompAnalyst/PayFactors) read as Salary.com CompAnalyst
Salary.com by CompAnalyst read as Salary.com CompAnalyst
Unresolved, counted raw
APEC
ARC-AGI
Artificial Analysis
Avasant
BDO private company executive compensation survey
BLS Occupational Employment and Wage Statistics OEWS
Betts Recruiting (Comp Engine)
Betts Recruiting Compensation Guide
Bureau of Labor Statistics (BLS) - OEWS
Bureau of Labor Statistics (BLS) – Occupational Employment and Wage Statistics (OEWS)
Bureau of Labor Statistics' free wage data
Comptryx
CrystalDiskMark
Deel Compensation
GAIA
GSM8K
Grant Thornton
HumanEval
Kienbaum
LINPACK (1979)
LinkedIn Job Postings
MGMA
MLCommons / MLPerf
MMLU
Mercer SME/market surveys
Mercer Total Remuneration Survey TRS / Management Benchmarking Database / Comptryx
OSWorld
PCMark
PassMark
Pavilion's B2B GTM Compensation Benchmarks
Payscale Knowledge / Payscale Salary Survey
RSM US
Radford McLagan
Radford, Aon
SPEC benchmarks
Sales Incentive Compensation Association (SICA)
Squirrel
Sullivan Cotter
U.S. Bureau of Labor Statistics (BLS) Occupational Employment Statistics
Whetstone (1964)
Xactly's Sales Compensation Report
Xactly's/Benchmarkit's B2B sales compensation survey
ravio.com
salarycube.com
Discontinued, still offered
No shut-down product was recommended here.