AI Token Counter
Count tokens in your prompt for GPT-4, Claude, Gemini, and more. Estimate API cost per request.
Enter Your Text / Prompt
Estimated Tokens
57
~4 chars per token
Context Used
0.0%
127.9K tokens left
Input Cost
$0.000285
at $0.0050/1K tokens
Total Est. Cost
$0.007785
input + 500 output tokens
Context Window Usage
0
128.0K tokens max
Estimate response length
Output cost: $0.007500
Total (input + output): $0.007785
Model Comparison
| Model | Context | Input $/1K | Output $/1K | Your text cost |
|---|---|---|---|---|
GPT-4o Selected | 128.0K | $0.0050 | $0.0150 | $0.000285 |
GPT-4 Turbo | 128.0K | $0.0100 | $0.0300 | $0.000570 |
GPT-4 | 8.2K | $0.0300 | $0.0600 | $0.001710 |
GPT-3.5 Turbo | 16.4K | $0.0005 | $0.0015 | $0.000029 |
Claude Opus 4 | 200.0K | $0.0150 | $0.0750 | $0.000855 |
Claude Sonnet 4 | 200.0K | $0.0030 | $0.0150 | $0.000171 |
Claude Haiku 4 | 200.0K | $0.0003 | $0.0013 | $0.000014 |
Gemini 2.5 Pro | 1000.0K | $0.0013 | $0.0100 | $0.000071 |
Gemini 2.0 Flash | 1000.0K | $0.0001 | $0.0004 | $0.000006 |
Llama 3.1 405B | 128.0K | $0.0050 | $0.0160 | $0.000285 |
Llama 3.1 70B | 128.0K | $0.0006 | $0.0008 | $0.000034 |
Prices as of May 2025. Token counts are estimates (~4 chars/token). Actual tokenization may vary ±10% depending on language and content type.
AI Token Counter — GPT, Claude & Gemini Token Estimator
Estimate token counts and API costs for any text across major AI models including GPT-4o, Claude Opus 4, Gemini 2.5 Pro, and Llama 3.1. Token estimation uses the widely applicable rule of approximately 4 characters per token for English text — the actual ratio from the cl100k_base tokenizer (GPT-4, Claude) and SentencePiece (Gemini). Code, URLs, and non-Latin scripts may tokenize differently. Context windows range from 8K tokens (GPT-4) to 1M tokens (Gemini 2.5 Pro). Pricing is per 1,000 tokens and varies significantly: GPT-3.5 Turbo inputs at $0.0005/1K vs GPT-4 at $0.03/1K. Always check the provider's pricing page for current rates.