LLM Token Budget Calculator
Calculate LLM API costs across current models: the monthly bill from tokens per request and traffic, and cheaper models in the same class.
Model
More options10% cached · 30 billable days
Cache hits are billed at $0.01/1M for this model (vs $0.1/1M input). Cache writes are not included.
Monthly cost on GPT-6 Luna
$8.87/ month
$0.000296 per request · 1,000 requests a day
DeepSeek V4 Flash would cost 56% less
Per month at your traffic. Tap a model to switch; check quality on your own task first.
- A day
- $0.2955
- A week
- $2.05
- Tokens per month
- 30M
List prices, standard tier: $0.1 in / $0.5 out per 1M tokens
Prices as of Oct 4, 2026; check the provider for current rates. Terms
How this works: Method, 5 sources,
How this works
Method
Cost per request = uncached input tokens × the model’s input price + cached input tokens × its cache-read price + output tokens × its output price, from each provider’s list prices per 1M tokens (cache writes are not included). Monthly cost = cost per request × requests a day × billable days, weekly = monthly ÷ (52 ÷ 12), and pasted text is estimated at 4 characters per token.
Sources
- OpenAI API pricing (OpenAI, prices checked 2026-10-04)
- Claude API pricing (Anthropic, prices checked 2026-10-04)
- Gemini API pricing (Google, prices checked 2026-10-04)
- Together AI pricing (Together AI, prices checked 2026-10-04)
- DeepInfra pricing (DeepInfra, prices checked 2026-10-04)
Limits
Prices as of Oct 4, 2026; check the provider for current rates.
Changes
- Added a printable report with the price comparisons and the cost per request.
- Prices updated to current provider lists; cache hits now use each model’s own cache-read price, and cheaper alternatives are compared within the same class of model.
- Results now update as you type, starting from a filled-in example.
- First published.
Frequently Asked Questions
What is a token in LLMs?
A token is a unit of text that LLMs process - typically 4 characters or 0.75 words on average. LLM providers charge based on the number of tokens processed (input + output). Understanding tokens helps you estimate and control your LLM API costs.
How do I count tokens in my prompt?
You can use online token counters (like this one), count characters and divide by 4 (rough estimate), or use libraries like tiktoken. This calculator accepts either direct token input or estimates from text length. For accuracy, use the official tokenizer for your provider.
What are cached tokens and why do they matter?
Cached tokens are input tokens that the provider has seen before in your conversation or system prompt. Cache hits are billed at a much lower rate that differs by provider and model: for example Anthropic charges 10% of the input price or less, OpenAI 5–10% on current models, and Google 10%. Writing to the cache can cost more than normal input (Anthropic and OpenAI charge 1.25× for cache writes). This calculator prices cache hits at each model's published cache-read rate; cache writes are not included.
Which LLM provider is the cheapest for my use case?
For simple tasks, small models like GPT-6 Luna, Gemini 3.5 Flash-Lite, DeepSeek V4 Flash or Llama 4 Scout are often 20-100x cheaper than frontier models. Mid-tier models such as GPT-6.1 Sol and Claude Sonnet 5.5 cost about the same as each other. This calculator compares cost across current models to help you find the best value — but check quality on your own task, since the cheapest model is not always good enough.
How do I reduce my LLM costs?
1. Use smaller, cheaper models when quality permits 2. Enable prompt caching 3. Reduce output token limits 4. Use compressed contexts 5. Implement retry logic with exponential backoff. The calculator's alternative model suggestions can help identify 10-90% savings opportunities.
What's the difference between input and output tokens?
Input tokens are what you send to the model (prompts, context, instructions). Output tokens are the generated response. Both are charged, but output tokens often cost more. For example, GPT-4o costs $2.50/1M input tokens but $10/1M output tokens. This calculator tracks both separately.