Mercury 2.5 Token Counter & Cost Calculator
Count tokens for Mercury 2.5 and estimate the exact API cost — $0.25
per 1M input tokens, $0.75 per 1M output tokens — using a real
o200k_base BPE tokenizer. Everything runs in your browser; nothing is uploaded.
Mercury 2.5
Counts are exact for OpenAI models (o200k_base is GPT-4o/4.1/5's tokenizer). For Claude, Gemini and other vendors — whose tokenizers are not public — treat the count as a close estimate (typically within ±15%).
Cost of this text across popular models
| Model | Input $/1M | Output $/1M | This text (input) | +500 output tokens |
|---|---|---|---|---|
| Claude Opus 5.5 (Adaptive Reasoning, Max Effort, Default Fallback) | 4.00 | 20.00 | $0 | $0 |
| Claude Fable 5.1 (Adaptive Reasoning, Max Effort, Default Fallback) | 10.00 | 50.00 | $0 | $0 |
| GPT-6 Astra (max) | 10.00 | 50.00 | $0 | $0 |
| Claude Opus 5 (Adaptive Reasoning, Max Effort) | 5.00 | 25.00 | $0 | $0 |
| Claude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback) | 10.00 | 50.00 | $0 | $0 |
| Muse Spark 1.3 (max) | 1.25 | 4.25 | $0 | $0 |
| GPT-6 Sol (max) | 2.00 | 10.00 | $0 | $0 |
| GPT-5.6 Sol (max) | 4.00 | 20.00 | $0 | $0 |
Close estimate (±15%): Inception does not publish its tokenizer, so this count uses OpenAI's o200k_base as an approximation.
Cost lookup — Mercury 2.5
| Tokens | Cost on Mercury 2.5 |
|---|---|
| 1K input tokens | 0.025¢ |
| 10K input tokens | 0.25¢ |
| 100K input tokens | $0.0250 |
| 1M input tokens | $0.2500 |
| 1K output tokens | 0.075¢ |
| 10K output tokens | 0.75¢ |
FAQ
How accurate is the Mercury 2.5 token count?
Close estimate (±15%): Inception does not publish its tokenizer, so this count uses OpenAI's o200k_base as an approximation. Paste your own text above to see the exact number for your prompt.
How is Mercury 2.5 API cost calculated?
Cost = (input tokens ÷ 1,000,000) × $0.25 + (output tokens ÷ 1,000,000) × $0.75 — Mercury 2.5's official per-1M-token prices from Artificial Analysis.
See full Mercury 2.5 pricing & benchmarks, or the main token counter for every model.