unitcostai

PR Review Cost

What does an AI PR review cost per pull request?

How this tool works

This calculator treats one pull-request review as the unit of analysis. It prices the model passes, diff and repository context, reranking, vector lookups, cache savings, infrastructure overhead, and fixed repo-index refresh needed to produce review output.

Use this when

  • You are packaging or evaluating AI-assisted PR review and need cost per reviewed pull request.
  • You want to separate review-pass count, diff context, repo retrieval, and refresh cost instead of using one hidden average.
  • You need a per-review chargeback or package-price floor before comparing vendor plans or rollout size.

How It Works

  1. Set the provider/model and expected model requests per PR review.
  2. Model diff tokens, retrieved repo context, reranking, vector lookups, cache, infra, and monthly repo refresh.
  3. Compare cost per PR review, monthly AI review spend, break-even review price, and margin.

Formulas

repo_refresh_share_usd = repo_refresh_monthly_usd / monthly_pr_reviews

cost_per_pr_review_usd = generation + retrieval + reranking + repo_refresh_share_usd + vector_db + cache + infra

optional_margin_pct = (review_price_or_chargeback_usd - cost_per_pr_review_usd) / review_price_or_chargeback_usd * 100

Assumptions and Units

  • Currency: USD
  • Unit of analysis: one reviewed pull request
  • Model requests per PR review includes retries, risk scans, and final review-comment generation
  • Repo refresh is amortized across monthly PR reviews before it appears in per-review cost
  • Pricing source: daily pricing snapshot in repo, no runtime scraping

Example Scenario

Start with three model passes per PR: one diff summary, one risk scan with retrieved repo context, and one final review comment pass. If the result is margin-safe, use AI Coding Plan Comparison or Codex vs Claude Cost to check whether the buying lane changes the package economics.

FAQs

Is this the same as AI coding-agent cost per developer? No. That page is per active developer-month. This page is per reviewed pull request.

What usually moves PR review cost first? Review passes, diff/context size, rerank depth, and repo refresh policy usually move faster than small model price differences.

Related resources: AI Coding Plan Comparison, Codex vs Claude Cost, AI Coding Agent Cost, PR Review Cost, AI Coding Agent Cost per Developer, Best AI Coding Plan for My Usage, Codex vs Claude Cost, Fixed Plan vs API Pricing for Coding Tools, Model Selection: Quality vs Unit Cost.

Pricing snapshot: 2026-04-24Provider: OpenAIModel: GPT-5.3 Codex

Decision Signal

Healthy

Current PR review margin is 75.7%. Use this to compare modeled review-agent cost with your per-review package price or internal chargeback.

Step 1 Provider and Model

Switch model assumptions using prices from the selected snapshot.

Step 2 Quick Mode

Use plain-language assumptions first. Open Advanced assumptions only if needed.
Starting pointApply the generic baseline or a more conservative downside scenario. If neither is selected, you are working from custom inputs.
Quick workflowStart with the closest workflow shape, then fine-tune the assumptions below.

Pull-request review workflows where diff size, review passes, repository context, and reranking determine cost per review.

Step 3 Advanced Assumptions

Tune retrieval, reranking, embeddings, vector, caching, and infra.
Show advanced inputs

Only adjust these once your Quick Mode assumptions feel realistic.

Scenario actions

Copy scenario URL

Paste into ChatGPT or Claude, or share with a teammate.

Save and track this scenario

Track pricing drift on this scenario and get an email if the latest result changes.

How tracking works

After you click Save and track, we carry this exact calculator state into the tracked-scenarios page so you can sign in and confirm the save.

We save your assumptions and the pricing snapshot used for this result.

When a newer pricing snapshot lands, we recompute the same scenario, show what changed, and email you if the latest result moved.

1 tracked scenario free, then $12/mo or $120/yr for up to 25 tracked scenarios.

Cost per PR review

$0.1215

Gross margin

75.7%

Estimated monthly review AI cost

$109.39

Estimated monthly review gross profit

$340.61

Top Cost Drivers

Most sensitive variables when each is moved up by 10%.
Requests Per User Month10.0%
Rerank Docs5.2%
Output Tokens2.1%

Totals

Summary metrics for monthly unit economics and margin.
Cost per model request
$0.04052
Cost per PR review
$0.1215
Gross margin %
75.7%
Break-even price
$0.1215
Cost per model request$0.04052
Cost per PR review$0.1215
Gross margin %75.7%
Break-even price$0.1215

Component Breakdown (USD/PR review)

Each cost component is computed independently and summed.

Largest cost block: reranking, not generation.

GenerationModel input/output token spend for requests.
$0.053
RetrievalExtra model input spend from retrieved context chunks.
$0.0109
RerankingReranker cost based on docs scored per request.
$0.072
Embeddings IngestionAmortized per-user share of the fixed monthly corpus embedding refresh cost.
$0
Vector DbVector database query cost across all requests.
$0.0001
CacheSavings from cache hits. Negative means lower total cost.
$-0.0166
InfraNon-model infra overhead per request.
$0.0021
GenerationModel input/output token spend for requests.$0.053
RetrievalExtra model input spend from retrieved context chunks.$0.0109
RerankingReranker cost based on docs scored per request.$0.072
Embeddings IngestionAmortized per-user share of the fixed monthly corpus embedding refresh cost.$0
Vector DbVector database query cost across all requests.$0.0001
CacheSavings from cache hits. Negative means lower total cost.$-0.0166
InfraNon-model infra overhead per request.$0.0021
Sensitivity RankingChange in total cost when one variable is increased by 10%.
VariableDelta cost %
Requests Per User MonthUser activity level per month.10.0%
Rerank DocsDocs reranked per request.5.2%
Output TokensGenerated tokens per request.2.1%
Input TokensPrompt-side tokens per request.1.7%
Cache Hit RateFraction of requests served by cache.-1.4%
Retrieved ChunksRetrieved chunk count per request.0.8%
Tokens Per ChunkAverage chunk size in tokens.0.8%
Vector Queries Per RequestVector query count per request.0.0%
Monthly Active UsersActive-user estimate used to amortize fixed monthly embedding refresh.-0.0%

Assumptions and Units

Explicit assumptions to keep outputs reproducible and auditable.
  • CurrencyUSD
  • Token unittoken
  • Pricing snapshot2026-04-24
  • Selected model rowOpenAI/GPT-5.3 Codex
  • Volume basisBusiness totals and fixed monthly terms use monthly PR reviews as the denominator
  • Embedding refreshAmortized per PR review from the fixed monthly repo/index refresh term
  • Cache componentNegative value means cost savings

Recommended Next Step

Use these links to lower top cost drivers without guessing.

Pressure-test the biggest review cost driver first. Compare coding-plan and model-routing choices only after review volume, diff context, and repo refresh assumptions are explicit.

Compare the buying lane next

AI Coding Plan ComparisonCodex vs Claude CostAI Coding Agent CostPR Review Cost

Compare infra providers

View Infra Recommendations

Sources and Snapshot

Pricing comes from the current dated snapshot.

Active Pricing Row

Selected model

OpenAI / GPT-5.3 Codex

  • Input tokens$1.75 / 1M
  • Output tokens$14 / 1M

Shared retrieval defaults

  • Embedding input$0.02 / 1M
  • Rerank docs$1 / 1K

Continue Analysis

Move to the next tool or guide without losing your current scenario.

Switch tools

Read guides