Free tools for building with LLMs
Seven calculators and builders for the questions that come up on every LLM project: what will it cost, will it fit, which model, and how do I phrase it. No sign-up, and what you type stays in your browser.
- LLM API cost calculatorPrice a workload across every model with a published rate. Set tokens, requests and caching; get a ranked monthly bill.
- Token counterPaste text to estimate tokens, see how it splits, get an exact count with OpenAI's tokenizer and price it per model.
- GPU and VRAM calculatorWork out the memory an open-weight model needs at each precision, including the KV cache, and which GPUs or Macs fit it.
- Which LLM should I use?Answer five questions about your task, budget and deployment. Get a shortlist of three models with the reasons.
- Context window visualizerSee what 8K, 128K or 1M tokens hold in pages, books, code and meeting hours, and compare model limits.
- Prompt builderAssemble a structured prompt from role, task, context, format and constraints. Copy it or download it.
- Self-host vs API calculatorFind the monthly volume where renting GPUs beats paying per token, with your own prices and throughput.
Which tool do you need?
Budgeting a feature? Start with the cost calculator: enter tokens per request and requests per day, and it ranks every priced model by monthly bill. If you don't know your token counts, paste a typical prompt into the token counter first.
Choosing a model? The model picker asks five questions and returns three candidates with the reasons. Put them side by side in the compare tool, then test them on your own data.
Thinking about self-hosting? Check whether a model fits your GPU or Mac with the VRAM calculator, then see whether renting GPUs beats paying per token at your volume with the break-even calculator.
Working with long inputs? The context window visualizer turns token limits into pages, books and hours of transcript. And when you're ready to write the prompt itself, the prompt builder assembles a structured one from your answers.
Building with LLMs? Get a vendor-neutral plan.
Tell us what you're building. Within 2 business days you get a model shortlist, an architecture outline and a cost estimate, plus optional introductions to vetted implementation partners.
- Free and vendor-neutral: no model provider pays for placement.
- Your details go only to partners you agree to.
Frequently asked questions
Are these LLM tools free?
Yes, all of them, with no account. They run entirely in your browser: the text you paste into the token counter or prompt builder isn't uploaded anywhere.
Where do the prices come from?
From each provider's official pricing page, checked on the date shown on the tool. Every model page lists the exact sources. Prices change often, so confirm on the provider's site before you commit to a budget.
How accurate is the VRAM calculator?
The weights figure is exact arithmetic on the parameter count and precision. The KV cache figure is an estimate that depends on each model's attention design, which the calculator accounts for where it's known. Leave headroom of 10-20% for real deployments.