AI Token + Pricing Calculator
Paste text to estimate its tokens and compare input and output cost across GPT, Claude, Gemini and other catalog models, with CSV export.
AI Token + Pricing Calculator
997 characters, about 250 tokens
Token estimates by tokenizer
Estimates based on average characters per token. Actual counts vary with content.
Cost comparison (for this text)
| Est. tokens | ||||
|---|---|---|---|---|
| GPT-OSS 20B (Groq)Cheapest | ~250 | $0.0000 | $0.0000 | $0.0000 |
| GPT-5-nano | ~250 | $0.0000 | $0.0001 | $0.0001 |
| GPT-4.1-nano | ~250 | $0.0000 | $0.0001 | $0.0001 |
| Gemini 2.5 Flash-Lite | ~250 | $0.0000 | $0.0001 | $0.0001 |
| GPT-6 Luna | ~250 | $0.0000 | $0.0001 | $0.0002 |
| Qwen3.8-Flash | ~250 | $0.0000 | $0.0001 | $0.0002 |
| Mistral Small 4 | ~250 | $0.0000 | $0.0001 | $0.0002 |
| GPT-OSS 120B (Groq) | ~250 | $0.0000 | $0.0001 | $0.0002 |
| Codestral 25.08 | ~250 | $0.0000 | $0.0002 | $0.0003 |
| GPT-5.6 Luna | ~250 | $0.0000 | $0.0003 | $0.0003 |
| GPT-5.4 Nano | ~250 | $0.0000 | $0.0003 | $0.0004 |
| DeepSeek V4.1 Flash | ~250 | $0.0000 | $0.0003 | $0.0004 |
| GPT-4.1-mini | ~250 | $0.0001 | $0.0004 | $0.0005 |
| Mistral Large 3 | ~250 | $0.0001 | $0.0004 | $0.0005 |
| GPT-5-mini | ~250 | $0.0000 | $0.0005 | $0.0006 |
| Gemini 3.5 Flash-Lite | ~250 | $0.0000 | $0.0006 | $0.0007 |
| Gemini 2.5 Flash | ~250 | $0.0000 | $0.0006 | $0.0007 |
| Grok Build 0.1 | ~250 | $0.0003 | $0.0005 | $0.0008 |
| Grok 4.3 | ~250 | $0.0003 | $0.0006 | $0.0009 |
| Grok 4.20 | ~250 | $0.0003 | $0.0006 | $0.0009 |
| Gemini 3.8 Flash | ~250 | $0.0002 | $0.0009 | $0.0011 |
| Gemini 3.7 Flash | ~250 | $0.0002 | $0.0009 | $0.0011 |
| Gemini 3.6 Flash | ~250 | $0.0002 | $0.0009 | $0.0011 |
| GPT-5.4 Mini | ~250 | $0.0002 | $0.0011 | $0.0013 |
| DeepSeek V4 Pro | ~250 | $0.0003 | $0.0010 | $0.0013 |
| o4-mini | ~250 | $0.0003 | $0.0011 | $0.0014 |
| o3-mini | ~250 | $0.0003 | $0.0011 | $0.0014 |
| Muse Spark 1.3 | ~250 | $0.0003 | $0.0011 | $0.0014 |
| Claude Haiku 4.5 | ~285 | $0.0003 | $0.0014 | $0.0017 |
| Grok 4.7 | ~250 | $0.0005 | $0.0015 | $0.0020 |
| Grok 4.6 | ~250 | $0.0005 | $0.0015 | $0.0020 |
| Grok 4.5 | ~250 | $0.0005 | $0.0015 | $0.0020 |
| Qwen3.8-Max | ~250 | $0.0005 | $0.0015 | $0.0020 |
| Mistral Medium 3.5 | ~250 | $0.0004 | $0.0019 | $0.0022 |
| o3 | ~250 | $0.0005 | $0.0020 | $0.0025 |
| GPT-4.1 | ~250 | $0.0005 | $0.0020 | $0.0025 |
| Gemini 3.5 Flash | ~250 | $0.0004 | $0.0023 | $0.0026 |
| GPT-5.1 | ~250 | $0.0003 | $0.0025 | $0.0028 |
| GPT-5 | ~250 | $0.0003 | $0.0025 | $0.0028 |
| Gemini 2.5 Pro | ~250 | $0.0003 | $0.0025 | $0.0028 |
| GPT-6.1 Sol | ~250 | $0.0005 | $0.0025 | $0.0030 |
| GPT-6 Sol | ~250 | $0.0005 | $0.0025 | $0.0030 |
| Claude Sonnet 5.5 | ~285 | $0.0006 | $0.0029 | $0.0034 |
| Claude Sonnet 5 | ~285 | $0.0006 | $0.0029 | $0.0034 |
| GPT-5.6 Terra | ~250 | $0.0005 | $0.0030 | $0.0035 |
| Gemini 3.1 Pro Preview | ~250 | $0.0005 | $0.0030 | $0.0035 |
| GPT-5.3-Codex | ~250 | $0.0004 | $0.0035 | $0.0039 |
| GPT-5.2 | ~250 | $0.0004 | $0.0035 | $0.0039 |
| GPT-5.4 | ~250 | $0.0006 | $0.0037 | $0.0044 |
| Claude Sonnet 4.6 | ~285 | $0.0009 | $0.0043 | $0.0051 |
| Claude Sonnet 4.5 | ~285 | $0.0009 | $0.0043 | $0.0051 |
| GPT-5.6 Sol | ~250 | $0.0010 | $0.0050 | $0.0060 |
| Claude Opus 5.5 | ~285 | $0.0011 | $0.0057 | $0.0068 |
| Claude Opus 5 | ~285 | $0.0014 | $0.0071 | $0.0085 |
| Claude Opus 4.8 | ~285 | $0.0014 | $0.0071 | $0.0085 |
| Claude Opus 4.7 | ~285 | $0.0014 | $0.0071 | $0.0085 |
| Claude Opus 4.6 | ~285 | $0.0014 | $0.0071 | $0.0085 |
| Claude Opus 4.5 | ~285 | $0.0014 | $0.0071 | $0.0085 |
| GPT-5.5 | ~250 | $0.0013 | $0.0075 | $0.0087 |
| GPT-6 Astra | ~250 | $0.0025 | $0.0125 | $0.0150 |
| Claude Fable 5.1 | ~285 | $0.0029 | $0.0142 | $0.0171 |
| Claude Fable 5 | ~285 | $0.0029 | $0.0142 | $0.0171 |
| o3-pro | ~250 | $0.0050 | $0.0200 | $0.0250 |
| GPT-5.2 Pro | ~250 | $0.0053 | $0.0420 | $0.0473 |
| GPT-5.5 Pro | ~250 | $0.0075 | $0.0450 | $0.0525 |
| GPT-5.4 Pro | ~250 | $0.0075 | $0.0450 | $0.0525 |
Prices per 1M tokens. Total = input + output cost for this text. Data as of October 3, 2026.
What this tool does
Paste any text into the AI Token and Pricing Calculator to count its tokens and compare what a call would cost across GPT, Claude, Gemini, and other models. It handles batch estimation for many inputs at once and exports results to CSV, all in your browser with no signup and nothing sent to a server.
Updated . Provided as is. Check the output before you rely on it in production.
How to use AI Token + Pricing Calculator
- 1
Paste text and count tokens across multiple LLMs
Paste your prompt or response text. The calculator estimates token counts from character-per-token ratios for the GPT, Claude, Gemini, Grok, DeepSeek and Mistral families, so you can compare relative token efficiency. The Token Counter gives exact o200k_base counts for OpenAI models.
- 2
Calculate API cost for your prompts
Paste prompt + expected response. The calculator counts tokens and multiplies by current model rates so you can estimate cost before deploying and keep API spend predictable.
- 3
Compare LLM costs across API providers
Run the same prompt across multiple models. The calculator shows costs side-by-side across providers so you can optimize for the right price/performance trade-off.
- 4
Estimate batch job costs before running
If processing 10,000 documents with summarization, calculator estimates total cost across models. Batch of 10k x 1,000 tokens average = 10M tokens. At $10 per 1M input tokens that is $100. Know budget before committing.
- 5
Sanity-check a suspicious bill
Re-enter the token counts you were billed for and compare the estimate against the invoice. A gap points at longer prompts or more verbose output than you assumed, which is where runaway loops show up.
Questions and answers
What is the AI Token + Pricing Calculator?
Is this tool free?
Does this tool store or send my text?
Which LLM is cheapest for my use case?
For AI agents: how to call this tool
Machine-readable contract, endpoints and examples. Humans can ignore this section.
Best Path For Builders
Browser workflow
Runs instantly in the browser with private local processing and copy/export-ready output.
Browser Workflow
This tool is optimized for instant in-browser execution with local data handling. Run it here and copy/export the output directly.
/ai-token-pricing-calculator/
For automation planning, fetch the canonical contract at /api/tool/ai-token-pricing-calculator.json.
Token Pricing Reference — October 3, 2026
Cost per 1 million tokens for 66 models. Use the calculator above to estimate costs for your specific workload.
| Model | Family | Input / 1M tokens | Output / 1M tokens | Cost per 1K tokens |
|---|---|---|---|---|
| Claude Fable 5.1 | Claude | $10.00 | $50.00 | $0.030 |
| Claude Opus 5.5 | Claude | $4.00 | $20.00 | $0.012 |
| Claude Opus 5 | Claude | $5.00 | $25.00 | $0.015 |
| Claude Sonnet 5.5 | Claude | $2.00 | $10.00 | $0.006 |
| Claude Sonnet 5 | Claude | $2.00 | $10.00 | $0.006 |
| Claude Haiku 4.5 | Claude | $1.00 | $5.00 | $0.003 |
| Claude Fable 5 | Claude | $10.00 | $50.00 | $0.030 |
| Claude Opus 4.8 | Claude | $5.00 | $25.00 | $0.015 |
| Claude Opus 4.7 | Claude | $5.00 | $25.00 | $0.015 |
| Claude Opus 4.6 | Claude | $5.00 | $25.00 | $0.015 |
| Claude Opus 4.5 | Claude | $5.00 | $25.00 | $0.015 |
| Claude Sonnet 4.6 | Claude | $3.00 | $15.00 | $0.009 |
| Claude Sonnet 4.5 | Claude | $3.00 | $15.00 | $0.009 |
| GPT-6 Astra | GPT | $10.00 | $50.00 | $0.030 |
| GPT-6.1 Sol | GPT | $2.00 | $10.00 | $0.006 |
| GPT-6 Sol | GPT | $2.00 | $10.00 | $0.006 |
| GPT-6 Luna | GPT | $0.10 | $0.50 | $0.000 |
| GPT-5.6 Sol | GPT | $4.00 | $20.00 | $0.012 |
| GPT-5.6 Terra | GPT | $2.00 | $12.00 | $0.007 |
| GPT-5.6 Luna | GPT | $0.20 | $1.20 | $0.001 |
| GPT-5.5 | GPT | $5.00 | $30.00 | $0.018 |
| GPT-5.5 Pro | GPT | $30.00 | $180.00 | $0.10 |
| GPT-5.4 | GPT | $2.50 | $15.00 | $0.009 |
| GPT-5.4 Pro | GPT | $30.00 | $180.00 | $0.10 |
| GPT-5.4 Mini | GPT | $0.75 | $4.50 | $0.003 |
| GPT-5.4 Nano | GPT | $0.20 | $1.25 | $0.001 |
| GPT-5.3-Codex | GPT | $1.75 | $14.00 | $0.008 |
| GPT-5.2 | GPT | $1.75 | $14.00 | $0.008 |
| GPT-5.2 Pro | GPT | $21.00 | $168.00 | $0.095 |
| GPT-5.1 | GPT | $1.25 | $10.00 | $0.006 |
| GPT-5 | GPT | $1.25 | $10.00 | $0.006 |
| GPT-5-mini | GPT | $0.25 | $2.00 | $0.001 |
| GPT-5-nano | GPT | $0.050 | $0.40 | $0.000 |
| o3 | GPT | $2.00 | $8.00 | $0.005 |
| o3-pro | GPT | $20.00 | $80.00 | $0.050 |
| o4-mini | GPT | $1.10 | $4.40 | $0.003 |
| o3-mini | GPT | $1.10 | $4.40 | $0.003 |
| GPT-4.1 | GPT | $2.00 | $8.00 | $0.005 |
| GPT-4.1-mini | GPT | $0.40 | $1.60 | $0.001 |
| GPT-4.1-nano | GPT | $0.10 | $0.40 | $0.000 |
| Gemini 3.8 Flash | Gemini | $0.75 | $3.75 | $0.002 |
| Gemini 3.7 Flash | Gemini | $0.75 | $3.75 | $0.002 |
| Gemini 3.6 Flash | Gemini | $0.75 | $3.75 | $0.002 |
| Gemini 3.5 Flash | Gemini | $1.50 | $9.00 | $0.005 |
| Gemini 3.5 Flash-Lite | Gemini | $0.30 | $2.50 | $0.001 |
| Gemini 3.1 Pro Preview | Gemini | $2.00 | $12.00 | $0.007 |
| Gemini 2.5 Pro | Gemini | $1.25 | $10.00 | $0.006 |
| Gemini 2.5 Flash | Gemini | $0.30 | $2.50 | $0.001 |
| Gemini 2.5 Flash-Lite | Gemini | $0.10 | $0.40 | $0.000 |
| DeepSeek V4.1 Flash | DeepSeek | $0.30 | $1.20 | $0.001 |
| DeepSeek V4 Pro | DeepSeek | $1.32 | $3.96 | $0.003 |
| Grok 4.7 | Grok | $2.00 | $6.00 | $0.004 |
| Grok 4.6 | Grok | $2.00 | $6.00 | $0.004 |
| Grok 4.5 | Grok | $2.00 | $6.00 | $0.004 |
| Grok 4.3 | Grok | $1.25 | $2.50 | $0.002 |
| Grok 4.20 | Grok | $1.25 | $2.50 | $0.002 |
| Grok Build 0.1 | Grok | $1.00 | $2.00 | $0.002 |
| Mistral Medium 3.5 | Mistral | $1.50 | $7.50 | $0.004 |
| Mistral Small 4 | Mistral | $0.15 | $0.60 | $0.000 |
| Mistral Large 3 | Mistral | $0.50 | $1.50 | $0.001 |
| Codestral 25.08 | Mistral | $0.30 | $0.90 | $0.001 |
| Muse Spark 1.3 | Muse | $1.25 | $4.25 | $0.003 |
| Qwen3.8-Max | Qwen | $2.00 | $6.00 | $0.004 |
| Qwen3.8-Flash | Qwen | $0.15 | $0.47 | $0.000 |
| GPT-OSS 120B (Groq) | GPT | $0.15 | $0.60 | $0.000 |
| GPT-OSS 20B (Groq) | GPT | $0.075 | $0.30 | $0.000 |
Quick Cost Estimates
A typical 1,000-word English document is roughly 1,300 tokens on GPT-family tokenizers (OpenAI's rule of thumb is 750 words per 1,000 tokens); every vendor tokenizes differently, so treat it as an estimate. Cost depends on provider, model tier, and output length:
- Flagship reasoning/coding models usually have the highest output-token costs.
- Balanced mid-tier models are often the best quality-to-cost trade-off for production apps.
- Fast and budget variants can be dramatically cheaper for high-volume pipelines.
- Open-weight and budget providers are often best for bulk summarization and classification.
- The table above is the source of truth for current per-model rates and families.
Prompt caching can reduce input costs by 75-90% for repeated prefixes. All Anthropic, OpenAI, and Google models support caching.