Chat History Token Growth Calculator
See how resending chat history inflates tokens and cost.
With the default inputs, 70,300 tokens sent by message 20 — Full history costs $0.0135/session · a 6-turn window costs $0.00904. 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 input tokens (full history)
- 70,300 tokens
- Session cost — full history
- $0.0135
- Session cost — 6-turn window
- $0.00904
- Saved by windowing
- $0.00450
In a chatbot, every turn usually resends the whole conversation — so cost grows with each message. See how many tokens you're sending by the last message and how much a sliding window would save.
How this is calculated
Because each turn usually resends the whole conversation, input tokens grow message by message. We sum the tokens sent across the session — system prompt + all prior turns + the new message at each step — so you see the total by the last message, and compare it to a sliding window that keeps only recent turns.
Is this a good result? What to do next
If your by-the-last-message token count is many times the first message, unbounded history is quietly inflating cost and latency. A sliding window or periodic summary usually removes most of that growth with little quality loss on typical chat.
Typical planning ranges
- Full-history cost growth
- roughly quadratic in message count
- Sliding window
- bounds tokens to the last N turns
- System prompt
- re-sent every turn unless cached
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
- Use a sliding window of recent turns instead of full history.
- Summarize older turns to compress the carried context.
- Cache the system prompt so it isn't re-billed each turn.
Common mistakes
- Costing one message and multiplying by message count.
- Leaving history unbounded on long sessions.
When to use a different approach
For a full monthly chatbot bill, use the chatbot cost calculator. To quantify summarization savings, use the summarization savings calculator.
Worked example (defaults)
With the default inputs above, here is the result:
- Total input tokens (full history)
- 70,300 tokens
- Session cost — full history
- $0.0135
- Session cost — 6-turn window
- $0.00904
- Saved by windowing
- $0.00450
- Resending full history each turn grows cost quadratically with message count.
Sources & references
Frequently asked questions
Why does chat cost grow over time?+
Each new turn typically includes the full prior conversation as input, so input tokens (and cost) grow with every message.
How do I control it?+
Use a sliding window of recent turns, summarize old history, or cache the system prompt.
Related calculators
Answering a real question?
This calculator powers these problem-solving guides: