Skip to content
parezenparezen
ProductHow it worksPricingDocsChangelog
Sign inTry the free AI bill X-ray

Parezen launches with the full feature set. Join the launch waitlist

The control plane for AI spend.

Spend less on every AI model. See every saving, line by line.

Parezen is designed to read the AI usage you already pay for, show which feature spends what, and find where a smaller model, a cache or a shorter prompt would do the same job, counted against your own baseline. None of that runs on this site yet.

Try the free AI bill X-raySee a sample report

Early access. The X-ray is planned and not live yet.

Sample data · not a customer

Overview

Spend

$21,600

Baseline

$28,200

Gap vs baseline

$6,600

AprMayJunJulAugSepGap $6,600

Audit finding

Support replies

gpt-4.1 on OpenAI. A smaller model would do this chat reply.

$2,100Estimated

Savings ledger

  • Support repliesEstimated$2,100
  • Docs searchRealised$1,200
  • Nightly evalVerified$1,500
  • Coding reviewEstimated$300
  • Invoice Q&ARealised$500
  • Catalog copyVerified$500
  • Ticket classifyEstimated$500

Gap vs baseline

$6,600

Gap $6,600

Savings ledger

  • Support repliesEstimated$2,100
  • Docs searchRealised$1,200
  • Nightly evalVerified$1,500
  • Coding reviewEstimated$300
  • Invoice Q&ARealised$500
  • Catalog copyVerified$500
  • Ticket classifyEstimated$500

Built for teams spending on OpenAI, Anthropic, Google and Mistral

  • OpenAIOpenAI
  • AnthropicAnthropic
  • GoogleGoogle
  • MistralMistral

Providers Parezen is built to work with, not customers.

  • Planned, not a live connection
  • Your baseline, not an industry average
  • Published prices. Pricing updated 2 Oct 2026.

Every request, sent to the model that does the job.

Parezen sits between your app and every provider. Each request goes to the cheapest model that still meets your quality bar, and every change is recorded.

Your appOpenAI SDKParezen gatewayOpenAIAnthropicGeminiMistral

Requested model Cheaper model that passed your quality bar

Routing decisions

Sample data

  • Support repliesgpt-4.1 › gpt-4.1-mini26% lower
  • Docs searchclaude-sonnet › claude-haiku22% lower
  • Nightly evalgemini-2.5-flash › gemini-2.5-flash-lite42% lower
  • Coding reviewgpt-4.1 › gpt-4.1-mini6% lower

Illustration on sample data. Percentages come from the sample month. Routing is planned, not live.

From one request to a verified saving.

PlannedEverything below is shown on sample data.

  1. 1

    Read the bill

    Parezen reads the usage you already pay for and splits it by feature, model and team. You see which part of the product spends what.

    Spend by feature, one month

    • Support replies$8,200
    • Docs search$5,400
    • Coding review$4,800
    • Nightly eval$3,600
    • Ticket classify$2,600
  2. 2

    Find the cheaper path

    For each feature it looks for a smaller model, a cache or a shorter prompt that does the same job, then tests it on a sample of real requests before anything changes.

    Support replies

    Now

    gpt-4.1

    $8,200

    ›

    Candidate

    gpt-4.1-mini

    $6,100

    Shadow test on real requestsPasses your bar
  3. 3

    Switch, inside your rules

    A change goes live only if it passed the test and your policy allows it. Budgets and hard stops still hold. You can review, enable or ignore each one.

    Routing policy, support route

    • Right-size support repliesShadow test passed
    • Monthly budget hard stop$9,000 for this route
    • Apply changes automaticallyOff: you approve each one
  4. 4

    Count it against your baseline

    Every saving is the gap between what the same work cost before and what it costs now, at the same prices. It becomes verified when the month matches the provider bill.

    Savings ledger line

    Baseline$8,200−Actual$6,100=Gap$2,100
    1. Estimated
    2. Realised
    3. Verified

    Verified when the month matches the provider bill within 2%. Only verified savings are billable.

Spend by feature, one month

  • Support replies$8,200
  • Docs search$5,400
  • Coding review$4,800
  • Nightly eval$3,600
  • Ticket classify$2,600

Sample data · not a customer

Example situations

Illustrative example, not a customer

A Series B support-AI team

About $40,000 a month across OpenAI and Anthropic.

An X-ray would look for a smaller model for routine replies.

Sample task line: $18,000 baseline, $12,400 later, gap $5,600.

Illustrative example, not a customer

A product docs team

About $12,000 a month across Anthropic.

An X-ray would look for prompt caching on a repeated prefix.

Sample task line: $7,200 baseline, $5,100 later, gap $2,100.

Illustrative example, not a customer

A nightly eval job

About $8,000 a month across Google.

An X-ray would look for trimming long prompts.

Sample task line: $3,100 baseline, $2,200 later, gap $900.

Where AI bills leak, and the ways to save.

PlannedEverything below is shown on sample data.

  • Expensive model, cheap work

    Routine replies priced like the hardest task.

  • Cache left off

    The same prefix billed on every call.

  • No baseline, no saving

    A lower bill still needs the same work.

Example savings path

High assumption, $20,000 a month. Overlap removed, so this is less than adding the lines.

Example reduction$4,724 (24%)

  1. $20,000
    Monthly spend
  2. −$2,240
    Right-size the model
  3. −$1,176
    Provider prompt caching
  4. −$543
    Output control
  5. −$540
    Batch async work
  6. −$225
    Coding-agent output
  7. $15,276
    After assumptions
Example only, from the estimator assumptions below. Not a result.

Ways to save

  • 8–16%

    Right-size the model

    Use a smaller model where it does the same job.

    0%20%
  • 4–10%

    Provider prompt caching

    Bill a repeated prefix once.

    0%20%
  • 2–6%

    Output control

    Stop where a short answer is enough.

    0%20%
  • 8–18%

    Batch async work

    Price work that can wait.

    0%20%
  • 3–8%

    Coding-agent output

    Shorter coding-agent output.

    0%20%
  • Opt-in

    Cross-provider switching

    Not in the example estimator. Only between providers you connect.

Ranges are example estimator assumptions, as a share of the affected spend. Not a result.

Full list of ways to save

See where the money actually goes.

PlannedEverything below is shown on sample data.

Switch the breakdown. The same sample month stays in view.

Spend breakdown

SpendBaseline

OpenAI: $10,600 spend, $13,000 baselineAnthropic: $4,200 spend, $5,400 baselineGoogle Gemini: $3,800 spend, $5,800 baselineMistral: $3,000 spend, $4,000 baseline
  • OpenAI$10,600$13,000
  • Anthropic$4,200$5,400
  • Google Gemini$3,800$5,800
  • Mistral$3,000$4,000
Show the table
Sample spend by provider
providerSpendBaseline
OpenAI$10,600$13,000
Anthropic$4,200$5,400
Google Gemini$3,800$5,800
Mistral$3,000$4,000
gpt-4.1: $10,600 spend, $13,000 baselineclaude-sonnet: $4,200 spend, $5,400 baselinegemini-2.5-flash: $2,100 spend, $3,600 baselinemistral-medium: $2,100 spend, $2,600 baselinegemini-2.5-pro: $1,700 spend, $2,200 baselinemistral-small: $900 spend, $1,400 baseline
  • gpt-4.1$10,600$13,000
  • claude-sonnet$4,200$5,400
  • gemini-2.5-flash$2,100$3,600
  • mistral-medium$2,100$2,600
  • gemini-2.5-pro$1,700$2,200
  • mistral-small$900$1,400
Show the table
Sample spend by model
modelSpendBaseline
gpt-4.1$10,600$13,000
claude-sonnet$4,200$5,400
gemini-2.5-flash$2,100$3,600
mistral-medium$2,100$2,600
gemini-2.5-pro$1,700$2,200
mistral-small$900$1,400
Support: $6,100 spend, $8,200 baselineEng: $4,500 spend, $4,800 baselineDocs: $4,200 spend, $5,400 baselineEval: $2,100 spend, $3,600 baselineOps: $2,100 spend, $2,600 baselineFinance: $1,700 spend, $2,200 baselineCatalog: $900 spend, $1,400 baseline
  • Support$6,100$8,200
  • Eng$4,500$4,800
  • Docs$4,200$5,400
  • Eval$2,100$3,600
  • Ops$2,100$2,600
  • Finance$1,700$2,200
  • Catalog$900$1,400
Show the table
Sample spend by project
projectSpendBaseline
Support$6,100$8,200
Eng$4,500$4,800
Docs$4,200$5,400
Eval$2,100$3,600
Ops$2,100$2,600
Finance$1,700$2,200
Catalog$900$1,400
Chat reply: $6,100 spend, $8,200 baselineCode review: $4,500 spend, $4,800 baselineRetrieval: $4,200 spend, $5,400 baselineBatch score: $2,100 spend, $3,600 baselineClassify: $2,100 spend, $2,600 baselineRAG: $1,700 spend, $2,200 baselineBatch draft: $900 spend, $1,400 baseline
  • Chat reply$6,100$8,200
  • Code review$4,500$4,800
  • Retrieval$4,200$5,400
  • Batch score$2,100$3,600
  • Classify$2,100$2,600
  • RAG$1,700$2,200
  • Batch draft$900$1,400
Show the table
Sample spend by task
taskSpendBaseline
Chat reply$6,100$8,200
Code review$4,500$4,800
Retrieval$4,200$5,400
Batch score$2,100$3,600
Classify$2,100$2,600
RAG$1,700$2,200
Batch draft$900$1,400
None: $15,000 spend, $19,800 baselinereview-bot: $4,500 spend, $4,800 baselineeval-bot: $2,100 spend, $3,600 baseline
  • None$15,000$19,800
  • review-bot$4,500$4,800
  • eval-bot$2,100$3,600
Show the table
Sample spend by agent
agentSpendBaseline
None$15,000$19,800
review-bot$4,500$4,800
eval-bot$2,100$3,600

What could you save? Try your numbers.

Example only. Not a promised result.

$20,000
$1,000$500,000
Providers in this example
Workload mix

Shares are scaled so they add up to 100%.

Advanced assumptions

Right-size the model

Provider prompt caching

Output control

Batch async work

Coding-agent output

Example estimate

$2,165–$4,724 per month

About 11–24% of $20,000 a month

  • Right-size the model$1,120–$2,240
  • Provider prompt caching$560–$1,400
  • Output control$220–$660
  • Batch async work$240–$540
  • Coding-agent output$90–$240

Levers overlap, so the combined range is less than the sum of the lines.

Plan band
Starter: $399/month, billed annually
Savings-share preview
$517 (cap $1,000)

Example only, based on published list prices and the assumptions above. Your X-ray replaces this with your real numbers.

Example estimate: $2,165–$4,724 per month

Every saving has a line. Every line has a baseline.

PlannedEverything below is shown on sample data.

  • Your own cost before the change.
  • The gap on the same work, same window.
  • The sample total is the sum of its lines.
Savings ledger

Estimated $2,900

Realised $1,700

Verified $2,000

Support replies

OpenAI · gpt-4.1

Estimated

$8,200 baseline → $6,100 · gap $2,100

Docs search

Anthropic · claude-sonnet

Realised

$5,400 baseline → $4,200 · gap $1,200

Nightly eval

Google Gemini · gemini-2.5-flash

Verified

$3,600 baseline → $2,100 · gap $1,500

Coding review

OpenAI · gpt-4.1

Estimated

$4,800 baseline → $4,500 · gap $300

Invoice Q&A

Google Gemini · gemini-2.5-pro

Realised

$2,200 baseline → $1,700 · gap $500

Catalog copy

Mistral · mistral-small

Verified

$1,400 baseline → $900 · gap $500

Ticket classify

Mistral · mistral-medium

Estimated

$2,600 baseline → $2,100 · gap $500

Sample savings ledger for September
FeatureProviderStatusBaselineSpendGap
Support repliesOpenAIEstimated$8,200$6,100$2,100
Docs searchAnthropicRealised$5,400$4,200$1,200
Nightly evalGoogle GeminiVerified$3,600$2,100$1,500
Coding reviewOpenAIEstimated$4,800$4,500$300
Invoice Q&AGoogle GeminiRealised$2,200$1,700$500
Catalog copyMistralVerified$1,400$900$500
Ticket classifyMistralEstimated$2,600$2,100$500
September total$28,200$21,600$6,600
Gap $6,600

Baseline $28,200Spend $21,600Gap $6,600

From your first audit to how you buy AI compute.

Roadmap. Nothing on this list is live yet.

  1. 01

    Audit and X-ray

    See where the money goes, from a usage export or a read-only connector.

  2. 02

    Verified savings

    Every saving counted against your baseline, then matched to the provider bill.

  3. 03

    Cost per task

    Agent and task budgets, priced per outcome instead of per token.

  4. 04

    FinOps and governance

    One place for spend, budgets, keys and policy across every provider.

  5. 05

    Self-tuning routing

    Routing that improves itself, inside the rules and quality bar you set.

  6. 06

    Buying AI compute

    Pooling demand so teams pay less for the same capacity. A later decision.

Your data, your call.

  • Run the gateway yourself

    The default. One container next to your app. Keys, logs and prompts stay on your servers.

    Planned

  • Hosted Quickstart

    For startups: change one URL to try it. Priced above running it yourself.

    Planned

  • Offline or air-gapped

    For regulated teams, in a later release.

    Planned

Read-only billing access. Prompt capture stays off unless you turn it on.

Security and privacy

Pricing

Free

Free

Under $1,000 a month of AI spend.

Team

$99/month, billed annually

For $1,000–$10,000 of AI spend.

Starter

$399/month, billed annually

For $10,000–$30,000 of AI spend.

Growth and up

From $1,599/month

From $30,000 a month of AI spend.

See full pricing

Pay a share only of savings we verify

  • No platform fee and no token markup.
  • 15% of verified savings for 12 months per change.
  • Capped at 5% of the pre-change spend of optimised workloads.
  • If verified savings are not above the fee, the share is $0 that month.
Full terms

The savings share is optional and sits on top of the plan price: 15% of verified savings for 12 months per change, capped at 5% of the pre-change spend of optimised workloads. If we don't save you more than we charge, you pay nothing that month. That waiver applies only to the savings share.

Parezen opens when the full feature set is ready. Join the launch waitlist

Try the free AI bill X-ray

Questions

What access do you need?

Read-only access, or a usage export that stays in this browser. Planned.

Does my traffic change?

No. A gateway that would switch a request is planned.

How is a saving counted?

The gap against your own baseline, on the same work. Invoice matching is planned.

Which providers can the X-ray read?

OpenAI, Anthropic, Google and Mistral. Providers Parezen is built to work with, not customers. Reading them is planned.

What does it cost?

Published prices. The X-ray is designed to be free. Paid plans are billed annually.

Where is my data stored?

This site does not store a provider key. A sample export stays in the browser.

See where your AI spend goes.

Try the free AI bill X-ray

Early access. The X-ray is planned and not live yet.

parezenparezen

Parezen reads the AI usage you already pay for and finds where you could spend less. A saving is the gap against your own baseline.

Product
  • Product
  • Pricing
  • Sample report
  • Changelog
Resources
  • Docs
  • Security
  • Calculator
  • Compare
Company
  • About
  • Contact
  • Enterprise
Legal
  • Privacy
  • Terms

Figures on this site are labelled sample data unless stated otherwise. List prices are the published prices. This site does not take a payment.

Contact [email protected].