Codebase Token Calculator
Estimate tokens and cost to feed a whole codebase to an LLM.
With the default inputs, ~409,091 tokens — 200 files · $0.8182 to send once to GPT-4.1. Enter your own numbers below to recompute instantly; the full step-by-step math is shown under the worked example.
Results update automatically as you type.
- Total characters
- 1,350,000
- Estimated tokens
- 409,091
- Input cost (once)
- $0.8182
Estimate how many tokens your codebase is and what it costs to send it to a model once. Set the number of files and their size to plan context for code assistants and review agents.
How this is calculated
We estimate total tokens as files × lines per file × characters per line ÷ ~3.3 (code's characters-per-token ratio), then price sending it once at your model's input rate. It's the quick way to see whether a repo could fit a context window and what a single pass costs.
Is this a good result? What to do next
Most real repos exceed even 1M-token windows, so the useful signal is usually 'this won't fit — index and retrieve instead.' If it does fit, weigh the one-pass cost against how often you'd resend it.
Typical planning ranges
- 1M-token models
- fit only small/medium repos
- 200 files × 150 lines
- ~400k tokens
- Typical line
- ~14 tokens
Ranges are typical planning figures to sanity-check your result, not authoritative benchmarks. Your numbers will vary with use case, volume, and vendor.
How to improve this number
- Index the repo and retrieve only relevant files per query.
- Exclude generated, vendored and test files from the context.
- Use a large-context model only when whole-repo context is essential.
Common mistakes
- Assuming a whole codebase fits in one prompt.
- Counting every file when only a subset is relevant.
When to use a different approach
For a single file or snippet, use the code token counter. To plan the full prompt layout, use the context window planner.
Worked example (defaults)
With the default inputs above, here is the result:
- Total characters
- 1,350,000
- Estimated tokens
- 409,091
- Input cost (once)
- $0.8182
- tokens ≈ 1,350,000 chars / 3.3 (code density)
Sources & references
Frequently asked questions
Can I fit a codebase in one prompt?+
Only small ones. Large repos exceed even 1M-token windows, so most tools index and retrieve relevant files instead of sending everything.
Related calculators
Answering a real question?
This calculator powers these problem-solving guides: