# AI Code Checker Pricing – Free Checks, Credit Packs & Plans | AI Code Checker

Canonical: https://aicodechecker.org/pricing

> Free to try, pay only for what you check. One credit is one check, and failed checks are refunded.

## Free tier, credit packs and subscriptions
- Free sign-up: $0 — 2 free credits on sign-up (valid for 30 days, no card required), sample reports open without an account, 7-day history retention.
- Credit unit: 1 credit checks up to 25,000 characters of one file (max 75,000 characters = 3 credits). Follow-up questions cost 1 credit. If every model fails, credits are refunded automatically.
- Starter pack: $4.99 one-time for 150 credits ($0.033/credit, never expires, unlocks batches up to 50 files and 12-month history).
- Standard pack: $15.99 one-time for 500 credits ($0.032/credit, never expires, unlocks batches up to 50 files and 12-month history).
- Large pack: $44.99 one-time for 1,500 credits ($0.030/credit, never expires, unlocks batches up to 50 files and 12-month history).
- Pro subscription: $9/month (300 credits/month, $0.030/credit) or $90/year (3,600 credits upfront, $0.025/credit). Includes 50-file batches and 12-month history; credits do not roll over.
- Business subscription: $29/month (1,200 credits/month, $0.024/credit) or $290/year (14,400 credits upfront, $0.020/credit). Includes 200-file batches, REST API access (/api/v1/*), and 24-month history; credits do not roll over.

## What every paid pack or plan unlocks
- Up to 75,000 chars per check (5x Free)
- Batch ZIP uploads up to 50 files
- Student folders, CSV & printable reports
- 1 credit = 1 check of one file. If every model fails, the credit is refunded. Prices in USD.

## Tool limits
- What it does: Estimates whether source code was written by an AI model or a human, returning a 0–100 score, verdict band, flagged line ranges with one-sentence reasons, and optional follow-up questions for the author.
- Score bands: 50–100 = "Likely AI-generated", 30–49 = "Worth a manual review", 0–29 = "No clear AI signals". Returns an insufficient-evidence note when code is too generic to judge.
- Supported languages: Python, JavaScript, TypeScript, Java, C++, C, Go, Rust, C#, PHP, Ruby, Kotlin.
- Input limits: At least 10 non-empty lines and up to 75,000 characters per file. Files over 25,000 characters are checked in segments of 25,000 characters (1 credit per segment, up to 3 segments).
- Follow-up questions: Generate author interview questions tied to specific lines (1 credit per report; automatically refunded on failure).
- Batch checking: Upload multiple files or a ZIP archive up to 50 files per batch on credit packs and Pro, or 200 files on Business. Supports per-student folder grouping (e.g., alice/main.py), CSV export, and printable PDF reports.
- REST API: Business subscribers can create API keys and automate checks via /api/v1/checks and /api/v1/credits.
- Benchmark (Nov 2026, 90 GitHub files across Python, Java, C++): Caught ~8 in 10 AI-written files (38/45 at score >=50) and wrongly flagged 1 in 45 human-written files.
- Retention & privacy: Code is sent to the model provider only for scoring and is never used for training. Reports and source code are kept for 7 days on free credits, 12 months on Pro or credit packs, and 24 months on Business; users can delete any check at any time.

Try the checker: https://aicodechecker.org/