EchoTuner
Client login Create your account
Create your account Client login

Monitoring

Deep monitoring: every AI model, checked

Every current model in ChatGPT, Claude, Gemini, Perplexity and Grok — 19 in all — because each one is built and tuned differently, and can say something different about you.

In short

Deep monitoring checks what AI says about your business at the model level: every current model people can use in each assistant — 19 in all, including fast, advanced, free and paid models. Each model is built and tuned differently, so two models from the same provider can give different answers about the same business. A Deep check includes everything in a Standard check and costs $4,000. Our recommendation is a quarterly Deep check on top of monthly Standard, which adds $1,500 per quarter.

Published by Echo Tuner · Last updated

Deep monitoring at a glance

Price$4,000 per check — includes the Standard providers
RecommendedQuarterly, on top of monthly Standard: +$1,500 per quarter
Other options$4,000/month (monthly) · $4,000 per quarter (quarterly on its own)
Models19 as of October 2026 — every current model people can use in each assistant, from fastest to most capable
IncludedEverything in Standard monitoring, plus the Audit & Plan
ContractNone — cancel anytime from your Plan tab; charges stop immediately

What does a Deep check add?

  • Every model, side by side — fast, advanced, free and paid versions in each assistant.
  • Where answers differ — which models mention you, which leave you out, and which cite competitors.
  • Everything in a Standard check — how often you appear at each depth of question, tone, sources, competitors and an updated plan.
  • A model list in every report, updated as providers release new models.

Which AI models does a Deep check cover?

Every current model people can use in each assistant — 19 as of October 2026 — shown by assistant from the fastest, lightest model to the most capable. Each one is built and tuned differently, so each can answer your customers' questions differently.

  1. OpenAI · ChatGPT

    Free users get GPT-5.6 Luna and paid plans GPT-5.6 Sol; GPT-6 Astra (called GPT-6 Pro in chat) is on the top plans. The newer GPT-6 Luna and GPT-6.1 Sol run in ChatGPT Work. Thinking levels and Pro versions are settings on these models.

    1. Fastest GPT-5.6 Luna Free default
    2. Fastest GPT-6 Luna ChatGPT Work
    3. Everyday GPT-5.6 Sol Paid default Standard check
    4. Advanced GPT-6.1 Sol ChatGPT Work
    5. Most capable GPT-6 Astra Top plans

  2. Anthropic · Claude

    Four Claude models. Free plans include Haiku and Sonnet, the default; paid plans add Opus and Fable, Anthropic's most capable model.

    1. Fastest Claude Haiku 4.5 Free and paid
    2. Everyday Claude Sonnet 5.5 The default Standard check
    3. Advanced Claude Opus 5.5 Paid plans
    4. Most capable Claude Fable 5.1 Paid plans

  3. Google · Gemini

    Free users get Gemini 3.6 Flash. Google AI Pro and Ultra subscribers get the newer Gemini 3.8 Flash, in the app and in AI Mode in Google Search. Gemini 3.1 Pro is the Pro option for harder questions.

    1. Fastest Gemini 3.6 Flash Free default
    2. Everyday Gemini 3.8 Flash Paid plans Standard check
    3. Most capable Gemini 3.1 Pro The Pro option

  4. Perplexity

    Perplexity's own Sonar models are built for search: every answer is based on live web results and cites its sources. Paid users can also pick other companies' models, which we check in those companies' own assistants.

    1. Fastest Sonar Quick answers
    2. Everyday Sonar Pro Research questions Standard check
    3. Advanced Sonar Reasoning Pro Multi-step searches
    4. Most capable Sonar Deep Research In-depth reports

  5. xAI · Grok

    The Grok app runs Grok 4.6 on every plan, with a Fast mode for quick answers and an Expert mode for harder ones; the Heavy plan runs several Grok 4.6 agents on one question at once. Grok 4.7, xAI's newest model (September 21, 2026), is in Grok Build and xAI's API.

    1. Everyday Grok 4.6 Free and paid Standard check
    2. Advanced Grok 4.7 Newest model
    3. Most capable Grok 4.6 Heavy Heavy plan
19 models in five AI assistants, as of October 2026. A Deep check asks every one; the five marked “Standard check” are the ones a Standard check asks.
Fastest
Quick and inexpensive to run. Often what free plans and quick answers use; they may search less and answer more briefly.
Everyday
The balance of speed and quality that many people use most of the time.
Advanced
Stronger reasoning for harder questions, usually on paid plans.
Most capable
The provider's most powerful model: slower, thinks longer, and often searches more before answering.

Why check every model?

  1. Models differEach model is trained and tuned differently, and some search the web more than others — so the same question can get a different answer on each.
  2. Your customers use different onesFree and paid users, and people on different apps, often get different models by default.
  3. Deep checks show the full pictureGaps that appear on only some models point to the facts and sources those models rely on.

Monthly or quarterly Deep checks?

Most businesses need a Deep check once a quarter; monthly Deep checks suit companies changing a lot at once.

  • Quarterly on top of monthly Standard (+$1,500 per quarter) — our recommendation.
  • Monthly — $4,000/month, Standard included: large changes under way, or a competitive market.
  • Quarterly on its own — $4,000 per quarter.

What Deep monitoring costs with each Standard choice

StandardDeepYou pay
MonthlyQuarterly$2,500/month + $1,500 per quarter
MonthlyMonthly$4,000/month (Standard included)
QuarterlyQuarterly$4,000 per quarter (Standard included)
NoneQuarterly$4,000 per quarter
NoneMonthly$4,000/month

How much does Deep monitoring cost?

$4,000 per check, which includes everything in a Standard check. Added once a quarter to monthly Standard, it costs $1,500 more per quarter, because it replaces that month's Standard check.

See it on the pricing page — no long-term contract.

Questions about Deep monitoring

Why do different AI models give different answers?

Each model is built and tuned differently, and models use web search differently, so they can name different businesses and cite different sources for the same question.

Which models do you check?

As of October 2026, 19 models — OpenAI: GPT-5.6 Luna, GPT-6 Luna, GPT-5.6 Sol, GPT-6.1 Sol, GPT-6 Astra; Anthropic: Claude Haiku 4.5, Claude Sonnet 5.5, Claude Opus 5.5, Claude Fable 5.1; Google: Gemini 3.6 Flash, Gemini 3.8 Flash, Gemini 3.1 Pro; Perplexity: Sonar, Sonar Pro, Sonar Reasoning Pro, Sonar Deep Research; xAI: Grok 4.6, Grok 4.7, Grok 4.6 Heavy.

These are the models people can use in each assistant today; retired models are left out. Providers release and retire models often, so your report lists every model tested in each check, and we update the list as the lineup changes.

Do I also need Standard monitoring?

No — a Deep check includes the Standard providers. Many businesses pair monthly Standard with a quarterly Deep check to see the full picture at a lower cost.

  • Standard monitoringOne check of each major AI provider, on hundreds of your customers' questions — monthly or quarterly.
  • Audit & PlanFind out what AI assistants say about your business today, why they say it, and exactly what to change first.
  • ReviewsWe prompt every customer to leave a review — with simple questions that bring out the details AI assistants read and repeat.

Sources

  1. OpenAI Help Center — GPT-5.6 and GPT-6 Pro in ChatGPT
    GPT-5.6 Luna is the default for Free and Go; GPT-5.6 Sol for paid plans; GPT-6 Pro is powered by GPT-6 Astra.
  2. ChatGPT documentation — Models in ChatGPT Work and Codex
    Astra, GPT-6.1 Sol and GPT-6 Luna; GPT-5.5 retires from ChatGPT on October 14, 2026.
  3. Claude Platform documentation — Models overview
    Claude Fable 5.1, Opus 5.5, Sonnet 5.5 and Haiku 4.5, from most capable to fastest.
  4. Claude Academy — Choosing the right Claude model
    Free plans include Haiku and Sonnet (the default); Pro and Max add Opus and Fable.
  5. Gemini API documentation — Models
    Gemini 3.8 Flash, Gemini 3.6 Flash and Gemini 3.1 Pro, among other models.
  6. Google — Introducing Gemini 3.8 Flash (September 2, 2026)
    Gemini 3.8 Flash is available to Google AI Pro and Ultra subscribers in the Gemini app and in AI Mode in Google Search.
  7. Gemini app release notes
    July 21, 2026: Gemini 3.6 Flash in the model menu for all users.
  8. Perplexity documentation — Sonar models
    Sonar, Sonar Pro, Sonar Reasoning Pro and Sonar Deep Research.
  9. xAI — Grok 4.7 (September 21, 2026)
    Grok 4.7 is available in Cursor, Grok Build and the Grok API.
  10. Fello AI — Grok pricing and plans (updated September 30, 2026)
    xAI's plan page describes the Grok app's Expert and Heavy modes as Grok 4.6; Grok 4.7 is the API flagship.

Next step

Add Deep monitoring to your plan.

Pick a plan or build your own on the pricing page — your total updates as you choose. No long-term contract.

Pricing Ask us a question