$ git diff main -- deployment/byok.md
commit a1b2c3d - author: Oliver Wakefield-Smith - verified 21 Jun 2026
BYOK (bring your own key) means pointing reviews at your own Claude / Gemini / OpenAI API key so token cost is paid directly to the model vendor.5 of the 19 tracked tools genuinely support BYOK: Qodo, Bito, Claude Code Review, Macroscope, Git AutoReview. Crossover math on /byok-breakeven.
@@ 5 tools that BYOK today @@genuine support, not 'enterprise quote'
1+Qodo - Per-user seat with pooled credits ($0.012/credit)src
2+Bito - Per-user seat (10x Developer) plus separate AI Code Review Agent seatsrc
3+Claude Code Review - Per-seat base ($100/seat/mo Team) plus token-metered usage on Anthropic APIsrc
4+Macroscope - Per-developer seat (Team), repriced June 2026src
5+Git AutoReview - Per-repository per month (not per seat)src
@@ who does NOT BYOK @@vendor owns the model bill
1-CodeRabbit
2-Greptile
3-Codacy
4-Sourcery
5-Graphite Reviewer
6-GitHub Copilot Code Review
7-Cursor BugBot
8-DeepSource
9-SonarQube
10-Snyk Code
11-Qlty
12-Augment Code Review
13-Launchable
14-Diffblue Cover
@@ model pricing for the BYOK path @@per 1M tokens, list
1
modelinputoutputsource
Claude Sonnet 4.5$3.00$15.00anthropic.com/pricing
Gemini 2.5 Pro$1.25$10.00ai.google.dev/pricing
GPT-4.1$2.00$8.00openai.com/api/pricing
@@ hidden BYOK ops cost @@add 10%
1-Key rotation: someone has to own the quarterly Anthropic / Google / OpenAI key roll
2-Rate limits: 429s and 529s under bursty PR loads need retry-with-backoff
3-Audit logging: per-developer usage attribution is your problem now
4-Cost attribution: token usage by repo / by team for chargeback
@@ cross-links @@related
1/byok-breakeven - per-tool crossover
2/tool/claude-code-review - the managed-BYOK hybrid
3/deployment/self-hosted