System Prompts
Browse owned catalog prompts, detect techniques, compare two, and open any prompt in the counter.
Compare two →Free AI
token counter
Count tokens and calculate costs for GPT-5.4, Claude, Gemini, Grok, DeepSeek and 29+ more AI models. Optimize your prompts, compare pricing, and ship — tokenization runs 100% in your browser, so your prompt content never touches our servers.
34 models8 providers100% private$0 free forever
Input tokens00 words · 0 chars · estimate
GPT-5.4$0.00in $0.0000 · out $0.0000 · 1:3
Context fill--400,000 window
Also in Fluxkit
Same weighing stack — catalog prompts and API shutdown tracker.
Deprecations
Track OpenAI, Anthropic, and Google model shutdowns with days left and migration snippets.
Open tracker →Models
Sort by price and context, compare up to four, then open a model page.
Open directory →API price list
Filter models →Input price, output price, and context window for 34 models. Open a row for the full rate card.
| Model | Provider | Input / 1M | Output / 1M | Context |
|---|---|---|---|---|
| GPT-5.4 | OpenAI | $2.5 | $15 | 400k context |
| GPT-5.4 mini | OpenAI | $0.75 | $4.5 | 400k context |
| GPT-5 | OpenAI | $1.25 | $10 | 400k context |
| GPT-5 Pro | OpenAI | $15 | $120 | 400k context |
| o3 | OpenAI | $10 | $40 | 200k context |
| o4-mini | OpenAI | $1.1 | $4.4 | 200k context |
| GPT-4.1 | OpenAI | $2 | $8 | 1.05M context |
| GPT-4.1 mini | OpenAI | $0.4 | $1.6 | 1.05M context |
| GPT-4o | OpenAI | $2.5 | $10 | 128k context |
| GPT-4o mini | OpenAI | $0.15 | $0.6 | 128k context |
| Claude Opus 4.7 | Anthropic | $5 | $25 | 1M context |
| Claude Opus 4.6 | Anthropic | $5 | $25 | 1M context |
| Claude Sonnet 4.6 | Anthropic | $3 | $15 | 1M context |
| Claude Sonnet 4.5 | Anthropic | $3 | $15 | 200k context |
| Claude Sonnet 5 | Anthropic | $2 | $10 | 1M context |
| Claude Haiku 4.5 | Anthropic | $1 | $5 | 200k context |
| Gemini 3.1 Pro | $2 | $12 | 1.05M context | |
| Gemini 3.1 Flash-Lite | $0.25 | $1.5 | 1.05M context | |
| Gemini 2.5 Pro | $1.25 | $10 | 1M context | |
| Gemini 2.5 Flash | $0.15 | $0.6 | 1M context | |
| Gemini 2.0 Flash | $0.1 | $0.4 | 1M context | |
| Grok 4.6 | xAI | $2 | $6 | 500k context |
| Grok 4.20 | xAI | $1.25 | $2.5 | 1M context |
| Grok 4.3 | xAI | $1.25 | $2.5 | 1M context |
| Grok 4.1 Fast | xAI | $0.2 | $0.5 | 2M context |
| DeepSeek V4 Pro | DeepSeek | $0.435 | $0.87 | 1M context |
| DeepSeek V4 Flash | DeepSeek | $0.14 | $0.28 | 1M context |
| DeepSeek R1 | DeepSeek | $0.55 | $2.19 | 128k context |
| Mistral Large | Mistral | $2 | $6 | 128k context |
| Mistral Small | Mistral | $0.1 | $0.3 | 128k context |
| Llama 4 Maverick | Meta | $0.27 | $0.85 | 1M context |
| Llama 3.3 70B | Meta | $0.59 | $0.79 | 128k context |
| Qwen3 235B | Alibaba | $0.2 | $0.6 | 262k context |
| Qwen 2.5 72B | Alibaba | $0.35 | $0.4 | 131k context |
How the counter works
Three steps from raw text to a planning cost estimate.
1. Paste your prompt
Single prompt or conversation. Variables like {{name}} resolve before counting.
2. Pick your model
34 models · 8 providers. Exact browser counts for GPT-family; estimates elsewhere.
3. Read the weight
Tokens, dollars, and fill update live. Optimize, compare, then export.
Learn how AI costs really work
All guides →ChatGPT Token Counter Guide: Budget Prompts Before You Hit Send
Learn what tokens are, why ChatGPT/Claude counts differ, and how to estimate prompt size privately in your browser with Fluxkit AI Lab.
How to Estimate OpenAI & Claude API Cost Before You Ship
A practical method to forecast LLM spend: measure input tokens, assume output size, multiply by requests, and sanity-check with Fluxkit’s cost tools.
GPT vs Claude Tokens: Why Counts Differ (and What To Do)
Understand why the same prompt shows different token estimates on GPT and Claude, and how to plan context and cost without guessing.
FAQ
Is Fluxkit free?
Yes. No signup, no API key for counting and planning estimates.
Do you store prompts on a server?
No. Work stays in the browser. Share links use the URL hash only.
How do I count tokens?
Paste text into the token counter. GPT-family models are counted in the browser. Other models are labeled estimates.
Is there an AI cost calculator?
Yes. The same prompt is priced across models. Open the model directory for input and output rates.
Where do prices come from?
Planning defaults. Confirm live rates on vendor pages — see model compare.