Gemini 3.8 Flash Pricing & Token Calculator
Paste a prompt to count its tokens and see what it costs on Gemini 3.8 Flash from Google, side by side with 16 other models. Prices checked October 15, 2026. Your text never leaves your browser.
Gemini 3.8 Flash API prices
gemini-3.8-flash
- Input
- $0.75 / 1M
- Cached input
- $0.075 / 1M
- Output
- $3.75 / 1M
- Context window
- 1.05M tokens
- A 1,000-token prompt with a 500-token reply costs $0.002625, or $2.62 for 1,000 such requests.
- Up to 1.05M input tokens and 66K output tokens per request.
- From January 1, 2027, Google has announced $1.50 input and $7.50 output per 1M tokens.
- Google has not published this model's tokenizer, so token counts here are estimates.
- Tokens
- 0
- Words
- 0
- Characters
- 0
Counted with OpenAI's o200k_base tokenizer, loaded when you start typing.
If you resend the same system prompt or document, most providers bill the repeated part at a lower cache price.
Cost for 1 request
AI API prices per 1 million tokens
Standard pay-as-you-go prices in US dollars, checked October 15, 2026. Batch and priority tiers cost differently.
| Model | Input | Cached | Output | Context | 1K in + 500 out |
|---|---|---|---|---|---|
| GPT-6 Astra | $10 | $1 | $50 | 1.05M | $0.035 |
| GPT-6.1 Sol | $2 | $0.10 | $10 | 1.05M | $0.007 |
| GPT-6 Luna | $0.10 | $0.01 | $0.50 | 1.05M | $0.00035 |
| GPT-5 | $1.25 | $0.125 | $10 | 400K | $0.00625 |
| GPT-5 mini | $0.25 | $0.025 | $2 | 400K | $0.00125 |
| GPT-5 nano | $0.05 | $0.005 | $0.40 | 400K | $0.00025 |
| GPT-4.1 mini | $0.40 | $0.10 | $1.60 | 1.05M | $0.0012 |
| GPT-4o mini | $0.15 | $0.075 | $0.60 | 128K | $0.00045 |
| Claude Fable 5.1 | $10 | $0.25 | $50 | 1M | $0.035 |
| Claude Opus 5.5 | $4 | $0.20 | $20 | 1M | $0.014 |
| Claude Sonnet 5.5 | $2 | $0.20 | $10 | 1M | $0.007 |
| Claude Haiku 5.5 | $0.10 | — | $0.50 | 1M | $0.00035 |
| Claude Haiku 4.5 | $1 | $0.10 | $5 | 200K | $0.0035 |
| Gemini 4 Argon price rises Dec 1 | $2 | $0.10 | $10 | 1.05M | $0.007 |
| Gemini 3.1 Pro (preview) | $2 | $0.20 | $12 | 1.05M | $0.008 |
| Gemini 3.8 Flash price rises Jan 1 | $0.75 | $0.075 | $3.75 | 1.05M | $0.002625 |
| Gemini 3.5 Flash-Lite | $0.30 | $0.03 | $2.50 | 1.05M | $0.00155 |
Sources: OpenAI pricing, Anthropic pricing, Google pricing. Always confirm on the provider's page before you budget.
Models with a similar price
About the AI Token & Cost Calculator
AI APIs bill by the token, and every model charges a different price for the text you send and the text it writes back. Paste a prompt to count its tokens, set how long the replies are and how many requests you make, and see the cost on each model, cheapest first. Prices come from each provider's pricing page and show the date they were last checked.
How to use it
- 1 Paste your prompt, system prompt, or document.
- 2 Set the reply length and the number of requests, and how much of the prompt is cached.
- 3 Compare the cost on each model, or open a model for its full price details.
Frequently asked questions
What is a token?
A token is a piece of text the model reads or writes: often a whole short word, part of a longer word, or a punctuation mark. In English, one token is about four characters or three quarters of a word. Bangla, Hindi, Chinese, and other scripts usually need more tokens for the same meaning.
Are the token counts exact?
They are exact for GPT-5, GPT-4.1, and GPT-4o, which use the public o200k_base tokenizer. Anthropic, Google, and the GPT-6 models have not published their tokenizers, so counts for those models are estimates and can differ by roughly 10 to 30 percent. Use the provider's token-counting API when you need an exact figure.
Why do output tokens cost more than input tokens?
Writing each output token takes a full pass through the model, while the prompt can be read in parallel. Providers price that extra work in, so output usually costs four to eight times as much as input.
What is prompt caching?
When you send the same long beginning of a prompt again, such as a system prompt or a document, providers can reuse their earlier work and bill that part at a much lower cached-input price. Set the cached share to see how much it saves.
Is my prompt sent to a server?
No. Tokens are counted in your browser, so your text never leaves your device.
Related tools
AI Output Cleaner
Clean ChatGPT and Claude text for Word, email, and WhatsApp
JSON Repair & LLM Output Validator
Fix broken JSON from ChatGPT, Claude, or Gemini and check it against a schema
AI Crawler Checker
See which AI crawlers, such as GPTBot and ClaudeBot, your robots.txt allows
llms.txt Generator
Create an llms.txt file that guides AI tools to your best pages
Speech to Text & Subtitle Maker
Transcribe audio and video to text and SRT subtitles in your browser
ChatGPT Table to Excel & CSV
Turn tables from ChatGPT, Claude, or Gemini into Excel and CSV files