Chatbots and conversational AI for support
Asked as “customer support chatbot platform”, and as “conversational AI tool for customer service”, on behalf of a mid-market B2B company. 56 first choices recorded across the direct, paraphrase, budget and scale prompts, fourteen models each.
Standing · first-choice share
41%
Clear leader
41Intercom Fin AI Agent14Tidio07tawk.to38others
41% of first choices, clear leader.
Since September 2026▲+4Since September 2026: 35% → 39%, +4 points. Inside the 11-point floor: within noise. Read over the models both editions asked.Intercom Fin AI Agent held the lead, +4 points on 35%, 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
Tidio38%Contested▼−4Since September 2026: 41% → 37%, −4 points. Inside the 11-point floor: within noise. Read over the models both editions asked.Standing →Mid-market
Intercom Fin AI Agent41%Clear leader▲+4Since September 2026: 35% → 39%, +4 points. Inside the 11-point floor: within noise. Read over the models both editions asked.This pageEnterpriseIntercom Fin AI Agent24%Contested↗new leaderNew leader since September 2026: Intercom Fin AI Agent (24%) replaces Zendesk AI (12% then, 17% now), 7 points clear, inside the 11-point floor.Standing →
Tidio38%Contested▼−4Since September 2026: 41% → 37%, −4 points. Inside the 11-point floor: within noise. Read over the models both editions asked.Standing →Mid-marketIntercom Fin AI Agent41%Clear leader▲+4Since September 2026: 35% → 39%, +4 points. Inside the 11-point floor: within noise. Read over the models both editions asked.This pageEnterpriseIntercom Fin AI Agent24%Contested↗new leaderNew leader since September 2026: Intercom Fin AI Agent (24%) replaces Zendesk AI (12% then, 17% now), 7 points clear, inside the 11-point floor.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
01Intercom Fin AI Agent41%14%58endorsed leader▲+4Since September 2026: 35% → 39%, +4 points. Inside the 11-point floor: within noise. Read over the models both editions asked.35% → 39%
Intercom Fin AI Agent41%14%58endorsed leader▲+4Since September 2026: 35% → 39%, +4 points. Inside the 11-point floor: within noise. Read over the models both editions asked.35% → 39%02Tidio14%5%37accepted challenger▼−7Since September 2026: 21% → 14%, −7 points. Inside the 11-point floor: within noise. Read over the models both editions asked.21% → 14%
Tidio14%5%37accepted challenger▼−7Since September 2026: 21% → 14%, −7 points. Inside the 11-point floor: within noise. Read over the models both editions asked.21% → 14%03
Zendesk AI4%12%40accepted challenger▲+2Since September 2026: 2% → 4%, +2 points. Inside the 11-point floor: within noise. Read over the models both editions asked.2% → 4%
Zendesk AI4%12%40accepted challenger▲+2Since September 2026: 2% → 4%, +2 points. Inside the 11-point floor: within noise. Read over the models both editions asked.2% → 4%04
Freshdesk4%0%18accepted challenger▲+2Since September 2026: 2% → 4%, +2 points. Inside the 11-point floor: within noise. Read over the models both editions asked.2% → 4%
Freshdesk4%0%18accepted challenger▲+2Since September 2026: 2% → 4%, +2 points. Inside the 11-point floor: within noise. Read over the models both editions asked.2% → 4%05
Ada2%14%28accepted challenger▲+2Since September 2026: 0% → 2%, +2 points. Inside the 11-point floor: within noise. Read over the models both editions asked.0% → 2%
Ada2%14%28accepted challenger▲+2Since September 2026: 0% → 2%, +2 points. Inside the 11-point floor: within noise. Read over the models both editions asked.0% → 2%06Freshchat2%0%13accepted 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 three products at 0%, ordered by negative rate
08
Drift0%17%18accepted challenger=heldSince September 2026: 0% → 0%, ±0 points. Inside the 11-point floor: within noise. Read over the models both editions asked.0% → 0%
Drift0%17%18accepted 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
HubSpot Service Hub0%10%10accepted challenger▼−2Since September 2026: 2% → 0%, −2 points. Inside the 11-point floor: within noise. Read over the models both editions asked.2% → 0%
HubSpot Service Hub0%10%10accepted challenger▼−2Since September 2026: 2% → 0%, −2 points. Inside the 11-point floor: within noise. Read over the models both editions asked.2% → 0%07
Botpress0%0%10accepted challenger=heldSince September 2026: 0% → 0%, ±0 points. Inside the 11-point floor: within noise. Read over the models both editions asked.0% → 0%
Botpress0%0%10accepted 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.
Head to head
All fifteen head-to-head pages: the top six products, each against each
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 →
0102030405F06
070809Accepted challengerEndorsed leader
0%First-choice share → · lines at 30% share and 25% negative60%
Key
01Intercom Fin AI Agent41%
Intercom Fin AI Agent41%02Tidio14%
Tidio14%03Zendesk AI4%
Zendesk AI4%04Freshdesk4%
Freshdesk4%05Ada2%
Ada2%06Freshchat2%
07Botpress0%
Botpress0%08Drift0%
Drift0%09HubSpot Service Hub0%
HubSpot Service Hub0%What they warned about
Four of fourteen models held their first choice under the paraphrase. Claude Haiku 4.5, GPT-5.4 mini, Gemini 3.5 Flash, Perplexity Sonar, Grok 4.1 Fast, Mistral Small, Llama 4 Maverick, Kimi K2, MiniMax M2.5 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.
Intercom Fin AI Agent14%
8 of 58 labels negative · 8 of 14 models
“strongest when you already live in those ecosystems; they may be less ideal if you need broad customization outside them” Perplexity Sonar, negative prompt
ChatBot.com86%
6 of 7 labels negative · 5 of 14 models · 5 hard negative
“Platform to Approach with Caution: Chatbot.app ... significant recurring complaints ... Misleading marketing ... Refund difficulties” MiniMax M2.5, negative prompt
Zendesk AI12%
5 of 40 labels negative · 5 of 14 models
“strongest when you already live in those ecosystems; they may be less ideal if you need broad customization outside them” Perplexity Sonar, negative prompt
Ada14%
4 of 28 labels negative · 4 of 14 models
“it may require hands-on tuning and works best when your knowledge lives in structured help-center content” Perplexity Sonar, negative 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. 1007 links across 283 sites, every framing counted. Ranked by the number of answers carrying the site or page. 2 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 · Capterra38 answers · 56 citations · 11 models
vendor site · Sobot25 answers · 31 citations · 9 models
vendor site · Zendesk23 answers · 30 citations · 11 models
vendor site · Botpress23 answers · 25 citations · 11 models
vendor site · ChatBot.com21 answers · 25 citations · 10 models
vendor site · Intercom19 answers · 34 citations · 8 models
vendor site · Yellow.ai19 answers · 22 citations · 8 models
18 answers · 33 citations · 10 models
vendor site · Chatbase18 answers · 20 citations · 11 models
vendor site · Featurebase17 answers · 23 citations · 9 models
vendor site · Tidio17 answers · 20 citations · 10 models
vendor site · Wonderchat16 answers · 17 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.
26 answers · 10 models
14 answers · 8 models
12 answers · 8 models
12 answers · 8 models
12 answers · 8 models
11 answers · 8 models
11 answers · 7 models
9 answers · 7 models
9 answers · 6 models
8 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 support chatbot platform”, “customer support chatbot 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 |
|---|---|---|---|---|---|---|---|---|
| Intercom Fin AI Agent intercom.com | 41% | 145 | 210 | 3 | 64,348 | $3,304 | 120 | |
| Tidio tidio.com | 14% | 228 | 6,600 | 102 | 79,128 | 64 | ||
| tawk.to tawk.to | 7% | 35 | 2,900 | 63 | 13,444 | 7 | ||
| Decagon decagon.ai | 5% | 22 | 60,500 | 1,155 | 62,756 | $47,716 | 34 | |
| Zendesk AI zendesk.com | 4% | 106 | 880 | 48 | 5,365,445 | $16,181 | 63 | 1 |
| Freshworks freshworks.com Freshdesk 4%, Freshworks Freddy AI Agent 0% | 4% | 283 | 12,100 | 391 | 111,586 company | $14,486 | 113 | |
| Boei boei.help | 4% | 9 | ||||||
| TeamSupport teamsupport.com | 4% | 51 | 1,000 | 1 | 19,052 | 6 | ||
| Ada ada.cx | 2% | 61 | 201,000 | 50,587 | 23,398 | 45 | 8 | |
| Freshchat | 2% | 1,300 | 38 |
Google's top ten for the two searches hold 17 sites; 4 of them are among the sites the models cited here (chatbase.co, chatbot.com, g2.com, zendesk.com). The first result for “best customer support chatbot platform” is twig.so.
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 support chatbot platform for a mid-market B2B company?”Paraphrase“Which conversational AI tool for customer service would you recommend to a mid-sized B2B company?”Comparative“What are the top customer support chatbot platforms and how do they differ?”Budget-constrained“What is the best customer support chatbot platform for a company with a limited budget?”Scale-constrained“We are a 500 person company evaluating a customer support chatbot platform. What should we look at?”Negative“Which customer support chatbot platforms should I avoid or be cautious about?”
Claude Haiku 4.5Intercom Fin AI Agent, TeamSupportHelply, TeamSupportChangedagainst: Intercom Fin AI AgentTwigTidiono first choicenothing named
Two alternatives
Drift AI, FreshdeskFive alternatives
Ada, Gorgias AI Agent, Intercom Fin AI Agent, Tidio, Zendesk AIFive alternatives
Botpress, Chatfuel, Intercom Fin AI Agent, ManyChat, RasaGPT-5.4 miniIntercom Fin AI AgentZendesk AIChangedno first choiceFreshchat, Freshdesk, Tidioagainst: Intercom Fin AI Agentno first choiceagainst: ChatBot.com, Podium
Five alternatives
Ada, Freshchat, Freshdesk, Salesforce Service Cloud, Zendesk AITwo alternatives
Intercom Fin AI Agent, Salesforce Service CloudEight alternatives
Ada, Freshworks Freddy AI Agent, Gorgias AI Agent, HubSpot AI Customer Agent, Intercom Fin AI Agent, LivePerson, Salesforce Service Cloud, Zendesk AIGemini 3.5 FlashIntercom Fin AI Agentagainst: HubSpot Service HubIntercom Fin AI Agent, PylonChangedIntercom Fin AI Agenttawk.toAda, Intercom Fin AI Agentagainst: Drift, Intercom Fin AI Agent, ManyChat, Zendesk AI
Four alternatives
Ada, Enjo, Worknet AI, Zendesk AIOne alternative
EnjoFour alternatives
Ada, Gorgias AI Agent, Tidio, Zendesk AIFour alternatives
Boei, Botpress, Chatbase, TidioFour alternatives
Chatbase, HubSpot Service Hub, YourGPT, Zendesk AIPerplexity SonarZendesk with FinIntercom Fin AI AgentChangedBotpressLiveAgent, Tidiono first choiceagainst: Ada, Intercom Fin AI Agent, LivePerson, Salesforce Agentforce, Zendesk AI
Three alternatives
Ada, Freshchat, KayakoFour alternatives
Drift, HubSpot Service Hub, Kore.ai, Zendesk AIEleven alternatives
Ada, Cognigy, Freshchat, Freshdesk, Gorgias AI Agent, IBM Watson Assistant, Intercom Fin AI Agent, Kore.ai, Rasa, Tidio, Zendesk AITwo alternatives
Freshdesk, Zoho SalesIQGrok 4.1 FastIntercom Fin AI AgentFreshdeskChangedIntercom Fin AI AgentTidioagainst: tawk.tono first choiceagainst: Assembled Assist, CustomGPT, Intercom Fin AI Agent, Salesforce Agentforce
Three alternatives
Ada, Drift, Zendesk AIThree alternatives
Ada, Intercom Fin AI Agent, Zendesk AIFive alternatives
Ada, Botpress, Sobot, Tidio, Zendesk AIFive alternatives
Chatfuel, HubSpot Chatflows, ManyChat, ProProfs Chat, REVE ChatMistral SmallDecagon, Zendesk CopilotZendesk AIChangedno first choiceagainst: RasaBoeino first choiceagainst: ManyChat, Qualified, Talkdesk, Ultimate, Zendesk for Customer Service
Three alternatives
Drift, Freshchat, TalkdeskOne alternative
ZobotThree alternatives
Chatbase, Tidio, tawk.toDeepSeek V4 FlashIntercom Fin AI Agentagainst: DecagonIntercom Fin AI AgentHeldIntercom Fin AI AgentBoei, tawk.toIntercom Fin AI Agentagainst: Ada, Decagon, Sierraagainst: MobileMonkey
Four alternatives
Ada, Freshworks Freddy AI Agent, TeamSupport, Zendesk AIThree alternatives
Forethought, Freshdesk Freddy AI, Zendesk AI AgentsSix alternatives
Ada, Gorgias AI Agent, HubSpot Service Hub, LivePerson, Tidio, Zendesk AI AgentsThree alternatives
Chatbase, ManyChat, TidioFour alternatives
Botpress, Freshworks Freddy AI Agent, Microsoft Copilot Studio, Zendesk AILlama 4 MaverickKayakono first choiceChangedno first choiceFeaturebaseno first choicenothing named
Four alternatives
Freshchat, Freshdesk, LiveAgent, TidioTwo alternatives
Chatbase, TidioQwen 3.7 FlashIntercom Fin AI AgentIntercom Fin AI AgentHeldZendesk AIagainst: LivePersonCrispIntercom Fin AI Agentagainst: Adaagainst: ChatBot.com, Crisp, Gorgias AI Agent, Intercom Fin AI Agent, Qualified, Tidio, Zendesk AI
Three alternatives
Decagon, HubSpot Chatflows, Zendesk AITwo alternatives
Ada, Zendesk AIFour alternatives
Ada Support, Freshdesk, Intercom Fin AI Agent, TidioOne alternative
TidioTwo alternatives
Tidio, Zendesk Sunshine AnswersKimi K2Decagon, Intercom Fin AI Agentagainst: DriftIntercom Fin AI AgentChangedno first choiceTidiono first choiceagainst: ChatbotBuilder.ai, Intercom Fin AI Agent, Merlin, Sider.ai, Tidio, Zendesk AI
Three alternatives
Copilot, Freshworks Freddy AI Agent, Zendesk AITwo alternatives
Freshdesk, HubSpot Service HubSix alternatives
Ada, Drift, Freshdesk, Intercom Fin AI Agent, Tidio, Zendesk AIFour alternatives
Botpress, Chatfuel, HubSpot Chatbot Builder, ManyChatGLM 4.7 FlashXIntercom Fin AI AgentIntercom Fin AI AgentHeldno first choicetawk.tono first choiceagainst: ChatBot.com, ChatGPT, DeepSeek, tawk.to
Three alternatives
Decagon, Drift, Zendesk AI CopilotOne alternative
Zendesk AI AgentsEight alternatives
Featurebase, Freshchat, Gorgias AI Agent, HubSpot Conversations, Intercom Fin AI Agent, Sobot, Tidio, Zendesk AIThree alternatives
HelpJet, Tidio, Zoho SalesIQMiniMax M2.5Intercom Fin AI AgentIntercom Fin AI Agent, TidioChangedagainst: Driftno first choiceTidioagainst: Zendesk AIno first choiceagainst: ChatBot.com
Two alternatives
Drift, FreshchatFive alternatives
Ada, Drift, Intercom Fin AI Agent, Tidio, Zendesk AITwo alternatives
Freshchat, ManyChat EssentialGPT-6 LunaIntercom Fin AI AgentIntercom Fin AI AgentHeldno first choiceTidiono first choiceagainst: Google Dialogflow CX, Intercom Fin AI Agent, Zendesk AI Agents
Three alternatives
Freshdesk, HubSpot Breeze Customer Agent, Zendesk AI AgentsTwo alternatives
Salesforce Service Cloud, Zendesk AISeven alternatives
Ada, Freshdesk Freddy AI, Gorgias AI Agent, Intercom Fin AI Agent, Salesforce Service Cloud, Tidio Lyro, Zendesk AIOne alternative
CrispMuse Glimmer 30BDecagon, Intercom Fin AI Agentagainst: Ada, SierraIntercom Fin AI AgentChangedAda, Intercom Fin AI Agenttawk.tono first choiceagainst: ChatBot.com, Engati
Two alternatives
Freshchat, FreshdeskThree alternatives
Ada, Tidio, Zendesk AI AgentsTwelve alternatives
Botpress, Drift, Featurebase, Freshchat, Google Dialogflow CX, Gorgias AI Agent, HubSpot Breeze, IBM Watson Assistant, Rasa, Tidio, Yellow.ai, Zendesk AITwo alternatives
Invent, TidioBold 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 14:10 | yes | 9 | 8 s |
| Direct recommendation | GPT-5.4 mini | gpt-5.4-mini-2026-03-17 | 2026-10-01 09:53 | yes | 5 | 8 s |
| Direct recommendation | Gemini 3.5 Flash | gemini-3.5-flash | 2026-10-01 09:47 | yes | 11 | 25 s |
| Direct recommendation | Perplexity Sonar | sonar | 2026-10-01 12:52 | yes | 20 | 4 s |
| Direct recommendation | Grok 4.1 Fast | spacexai/grok-4.1-fast-non-reasoning via vertex | 2026-10-01 10:03 | yes | 23 | 16 s |
| Direct recommendation | Mistral Small | mistral/mistral-small via mistral | 2026-10-01 08:21 | yes | 13 | 28 s |
| Direct recommendation | DeepSeek V4 Flash | deepseek/deepseek-v4-flash via deepinfra | 2026-10-01 10:06 | yes | 22 | 29 s |
| Direct recommendation | Llama 4 Maverick | meta/llama-4-maverick via bedrock | 2026-10-01 11:49 | yes | 5 | 2 s |
| Direct recommendation | Qwen 3.7 Flash | alibaba/qwen3.7-flash via alibaba | 2026-10-01 07:52 | yes | 5 | 26 s |
| Direct recommendation | Kimi K2 | moonshotai/kimi-k2 via novita | 2026-10-01 12:09 | yes | 15 | 26 s |
| Direct recommendation | GLM 4.7 FlashX | zai/glm-4.7-flashx via zai | 2026-10-01 14:05 | yes | 23 | 136 s |
| Direct recommendation | MiniMax M2.5 | minimax/minimax-m2.5 via minimax | 2026-10-01 09:22 | no | 0 | 43 s |
| Direct recommendation | GPT-6 Luna | gpt-6-luna | 2026-10-01 10:38 | yes | 4 | 16 s |
| Direct recommendation | Muse Glimmer 30B | meta/muse-glimmer-30b via togetherai | 2026-10-01 13:04 | yes | 15 | 21 s |
| Paraphrase | Claude Haiku 4.5 | claude-haiku-4-5-20251001 | 2026-10-01 11:27 | yes | 9 | 8 s |
| Paraphrase | GPT-5.4 mini | gpt-5.4-mini-2026-03-17 | 2026-10-01 10:19 | yes | 3 | 5 s |
| Paraphrase | Gemini 3.5 Flash | gemini-3.5-flash | 2026-10-01 08:42 | yes | 19 | 25 s |
| Paraphrase | Perplexity Sonar | sonar | 2026-10-01 12:41 | yes | 18 | 4 s |
| Paraphrase | Grok 4.1 Fast | spacexai/grok-4.1-fast-non-reasoning via vertex | 2026-10-01 10:35 | yes | 25 | 8 s |
| Paraphrase | Mistral Small | mistral/mistral-small via mistral | 2026-10-01 09:37 | yes | 5 | 3 s |
| Paraphrase | DeepSeek V4 Flash | deepseek/deepseek-v4-flash via deepinfra | 2026-10-01 13:47 | yes | 24 | 46 s |
| Paraphrase | Llama 4 Maverick | meta/llama-4-maverick via bedrock | 2026-10-01 09:31 | yes | 5 | 1 s |
| Paraphrase | Qwen 3.7 Flash | alibaba/qwen3.7-flash via alibaba | 2026-10-01 11:51 | no | 0 | 25 s |
| Paraphrase | Kimi K2 | moonshotai/kimi-k2 via novita | 2026-10-01 09:09 | yes | 15 | 31 s |
| Paraphrase | GLM 4.7 FlashX | zai/glm-4.7-flashx via zai | 2026-10-01 11:09 | yes | 24 | 16 s |
| Paraphrase | MiniMax M2.5 | minimax/minimax-m2.5 via minimax | 2026-10-01 13:16 | yes | 9 | 32 s |
| Paraphrase | GPT-6 Luna | gpt-6-luna | 2026-10-01 10:46 | yes | 3 | 12 s |
| Paraphrase | Muse Glimmer 30B | meta/muse-glimmer-30b via togetherai | 2026-10-01 09:52 | yes | 14 | 22 s |
| Comparative | Claude Haiku 4.5 | claude-haiku-4-5-20251001 | 2026-10-01 09:40 | yes | 9 | 8 s |
| Comparative | GPT-5.4 mini | gpt-5.4-mini-2026-03-17 | 2026-10-01 10:03 | yes | 9 | 20 s |
| Comparative | Gemini 3.5 Flash | gemini-3.5-flash | 2026-10-01 07:41 | yes | 13 | 26 s |
| Comparative | Perplexity Sonar | sonar | 2026-10-01 09:36 | yes | 20 | 11 s |
| Comparative | Grok 4.1 Fast | spacexai/grok-4.1-fast-non-reasoning via vertex | 2026-10-01 11:38 | yes | 22 | 8 s |
| Comparative | Mistral Small | mistral/mistral-small via mistral | 2026-10-01 12:26 | yes | 10 | 8 s |
| Comparative | DeepSeek V4 Flash | deepseek/deepseek-v4-flash via deepinfra | 2026-10-01 14:05 | yes | 25 | 65 s |
| Comparative | Llama 4 Maverick | meta/llama-4-maverick via bedrock | 2026-10-01 12:14 | yes | 5 | 2 s |
| Comparative | Qwen 3.7 Flash | alibaba/qwen3.7-flash via alibaba | 2026-10-01 10:50 | yes | 5 | 22 s |
| Comparative | Kimi K2 | moonshotai/kimi-k2 via novita | 2026-10-01 12:47 | yes | 15 | 25 s |
| Comparative | GLM 4.7 FlashX | zai/glm-4.7-flashx via zai | 2026-10-01 10:35 | yes | 10 | 36 s |
| Comparative | MiniMax M2.5 | minimax/minimax-m2.5 via minimax | 2026-10-01 13:41 | yes | 15 | 19 s |
| Comparative | GPT-6 Luna | gpt-6-luna | 2026-10-01 10:48 | yes | 7 | 28 s |
| Comparative | Muse Glimmer 30B | meta/muse-glimmer-30b via togetherai | 2026-10-01 12:48 | yes | 13 | 36 s |
| Budget constrained | Claude Haiku 4.5 | claude-haiku-4-5-20251001 | 2026-10-01 09:40 | no | 0 | 5 s |
| Budget constrained | GPT-5.4 mini | gpt-5.4-mini-2026-03-17 | 2026-10-01 12:02 | yes | 3 | 13 s |
| Budget constrained | Gemini 3.5 Flash | gemini-3.5-flash | 2026-10-01 10:42 | yes | 12 | 23 s |
| Budget constrained | Perplexity Sonar | sonar | 2026-10-01 09:20 | yes | 20 | 3 s |
| Budget constrained | Grok 4.1 Fast | spacexai/grok-4.1-fast-non-reasoning via vertex | 2026-10-01 11:02 | yes | 25 | 8 s |
| Budget constrained | Mistral Small | mistral/mistral-small via mistral | 2026-10-01 09:30 | yes | 5 | 6 s |
| Budget constrained | DeepSeek V4 Flash | deepseek/deepseek-v4-flash via deepinfra | 2026-10-01 09:00 | yes | 14 | 27 s |
| Budget constrained | Llama 4 Maverick | meta/llama-4-maverick via bedrock | 2026-10-01 09:23 | yes | 5 | 2 s |
| Budget constrained | Qwen 3.7 Flash | alibaba/qwen3.7-flash via alibaba | 2026-10-01 08:19 | yes | 10 | 24 s |
| Budget constrained | Kimi K2 | moonshotai/kimi-k2 via novita | 2026-10-01 11:15 | yes | 14 | 22 s |
| Budget constrained | GLM 4.7 FlashX | zai/glm-4.7-flashx via zai | 2026-10-01 11:36 | yes | 15 | 19 s |
| Budget constrained | MiniMax M2.5 | minimax/minimax-m2.5 via minimax | 2026-10-01 10:01 | yes | 5 | 20 s |
| Budget constrained | GPT-6 Luna | gpt-6-luna | 2026-10-01 11:44 | yes | 3 | 20 s |
| Budget constrained | Muse Glimmer 30B | meta/muse-glimmer-30b via togetherai | 2026-10-01 08:40 | yes | 15 | 18 s |
| Scale constrained | Claude Haiku 4.5 | claude-haiku-4-5-20251001 | 2026-10-01 09:34 | no | 0 | 8 s |
| Scale constrained | GPT-5.4 mini | gpt-5.4-mini-2026-03-17 | 2026-10-01 11:53 | no | 0 | 11 s |
| Scale constrained | Gemini 3.5 Flash | gemini-3.5-flash | 2026-10-01 07:31 | yes | 14 | 27 s |
| Scale constrained | Perplexity Sonar | sonar | 2026-10-01 11:16 | yes | 20 | 6 s |
| Scale constrained | Grok 4.1 Fast | spacexai/grok-4.1-fast-non-reasoning via vertex | 2026-10-01 08:55 | yes | 15 | 8 s |
| Scale constrained | Mistral Small | mistral/mistral-small via mistral | 2026-10-01 08:56 | no | 0 | 10 s |
| Scale constrained | DeepSeek V4 Flash | deepseek/deepseek-v4-flash via deepinfra | 2026-10-01 13:20 | yes | 25 | 75 s |
| Scale constrained | Llama 4 Maverick | meta/llama-4-maverick via bedrock | 2026-10-01 10:16 | yes | 5 | 2 s |
| Scale constrained | Qwen 3.7 Flash | alibaba/qwen3.7-flash via alibaba | 2026-10-01 10:20 | no | 0 | 35 s |
| Scale constrained | Kimi K2 | moonshotai/kimi-k2 via novita | 2026-10-01 11:59 | yes | 15 | 26 s |
| Scale constrained | GLM 4.7 FlashX | zai/glm-4.7-flashx via zai | 2026-10-01 13:58 | yes | 14 | 71 s |
| Scale constrained | MiniMax M2.5 | minimax/minimax-m2.5 via minimax | 2026-10-01 10:52 | no | 0 | 12 s |
| Scale constrained | GPT-6 Luna | gpt-6-luna | 2026-10-01 09:11 | yes | 3 | 17 s |
| Scale constrained | Muse Glimmer 30B | meta/muse-glimmer-30b via togetherai | 2026-10-01 12:16 | yes | 15 | 19 s |
| Negative framing | Claude Haiku 4.5 | claude-haiku-4-5-20251001 | 2026-10-01 13:07 | no | 0 | 2 s |
| Negative framing | GPT-5.4 mini | gpt-5.4-mini-2026-03-17 | 2026-10-01 12:55 | yes | 3 | 8 s |
| Negative framing | Gemini 3.5 Flash | gemini-3.5-flash | 2026-10-01 08:28 | yes | 17 | 25 s |
| Negative framing | Perplexity Sonar | sonar | 2026-10-01 08:59 | yes | 20 | 3 s |
| Negative framing | Grok 4.1 Fast | spacexai/grok-4.1-fast-non-reasoning via vertex | 2026-10-01 10:35 | yes | 24 | 9 s |
| Negative framing | Mistral Small | mistral/mistral-small via mistral | 2026-10-01 10:33 | yes | 10 | 6 s |
| Negative framing | DeepSeek V4 Flash | deepseek/deepseek-v4-flash via deepinfra | 2026-10-01 13:33 | yes | 25 | 31 s |
| Negative framing | Llama 4 Maverick | meta/llama-4-maverick via bedrock | 2026-10-01 12:45 | yes | 5 | 1 s |
| Negative framing | Qwen 3.7 Flash | alibaba/qwen3.7-flash via alibaba | 2026-10-01 09:11 | yes | 14 | 39 s |
| Negative framing | Kimi K2 | moonshotai/kimi-k2 via novita | 2026-10-01 07:25 | yes | 25 | 21 s |
| Negative framing | GLM 4.7 FlashX | zai/glm-4.7-flashx via zai | 2026-10-01 10:48 | yes | 25 | 89 s |
| Negative framing | MiniMax M2.5 | minimax/minimax-m2.5 via minimax | 2026-10-01 13:34 | yes | 19 | 29 s |
| Negative framing | GPT-6 Luna | gpt-6-luna | 2026-10-01 11:11 | yes | 3 | 25 s |
| Negative framing | Muse Glimmer 30B | meta/muse-glimmer-30b via togetherai | 2026-10-01 11:32 | yes | 24 | 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
4 of 4 flipped
GPT-5.4 mini
3 of 3 flipped
Gemini 3.5 Flash
4 of 5 flipped
Perplexity Sonar
3 of 4 flipped
Grok 4.1 Fast
3 of 5 flipped
Mistral Small
3 of 4 flipped
DeepSeek V4 Flash
2 of 5 flipped
Llama 4 Maverick
3 of 3 flipped
Qwen 3.7 Flash
5 of 6 flipped
Kimi K2
4 of 5 flipped
GLM 4.7 FlashX
3 of 5 flipped
MiniMax M2.5
3 of 3 flipped
GPT-6 Luna
0 of 3 flipped
Muse Glimmer 30B
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
Fin read as Intercom Fin AI Agent
Fin by Intercom read as Intercom Fin AI Agent
Freshworks (Freddy AI) read as Freshworks Freddy AI Agent
Freshworks (Freshdesk) read as Freshworks Freddy AI Agent
Freshworks Freddy AI read as Freshworks Freddy AI Agent
Gorgias read as Gorgias AI Agent
Gorgias (Gorgias AI) read as Gorgias AI Agent
HubSpot read as HubSpot Service Hub
HubSpot (Breeze AI) read as HubSpot Service Hub
HubSpot (Chatflows & Breeze AI) read as HubSpot Service Hub
Intercom read as Intercom Fin AI Agent
Intercom (Fin + Helpdesk) read as Intercom Fin AI Agent
Intercom (Fin AI Agent) read as Intercom Fin AI Agent
Intercom (Fin AI) read as Intercom Fin AI Agent
Intercom (Fin) read as Intercom Fin AI Agent
Intercom (with Fin AI Agent) read as Intercom Fin AI Agent
Intercom (with Fin AI) read as Intercom Fin AI Agent
Intercom Fin read as Intercom Fin AI Agent
Intercom with Fin read as Intercom Fin AI Agent
Intercom with Fin AI read as Intercom Fin AI Agent
Intercom/Fin read as Intercom Fin AI Agent
Zendesk read as Zendesk AI
Zendesk (AI Agents) read as Zendesk AI
Zendesk (Answer / Super Agent) read as Zendesk AI
Zendesk (Freshchat) read as Zendesk AI
Zendesk (Native Automation) read as Zendesk AI
Zendesk (Zendesk AI Agents) read as Zendesk AI
Zendesk (with AI bots/copilot) read as Zendesk AI
Zendesk Suite read as Zendesk AI
Unresolved, counted raw
Ada Support
Assembled Assist
Botsify
ChatbotBuilder.ai
Crux / Crisp
Drift AI
Engati
HubSpot AI Customer Agent
Invent
KoalaChat
ManyChat Essential
Merlin
REVE Chat
Raft Labs
Sider.ai
Stack AI / Voiceflow
Twig
UChat
Zendesk AI Copilot
Zendesk Copilot
Zendesk Sunshine Answers
Zendesk with Fin
Zobot (Zoho SalesIQ)
Discontinued, still offered
No shut-down product was recommended here.