Guides
Guides
Practical guides to choosing and using AI models: comparing them on your own work, what a context window is, prompting, and what models cost.
Using AI models well
- AI prompt library
A library of prompts for writing, summarizing, coding, data, learning and research: copy one, fill in the brackets, and use it with any model in llmwise.
- AI usage limits, explained: what each $20 plan tells you
What ChatGPT Plus, Claude Pro, Google AI Pro and Perplexity Pro say about their limits, sourced and dated, beside llmwise's count for every model.
- How to compare LLM models
Compare LLMs on your own questions: pick real tasks, run them on two or three models in one chat, and judge the answers on what matters to you.
- How to use Claude: through the API, or in a chat
Two ways to use Claude: Anthropic's API from code, or a chat. How to choose, which Claude model fits what, what each costs, and using Claude in llmwise.
- How we write and check these pages
Who writes llmwise's guides and comparisons, how AI drafts them from our own data, how every fact is sourced, and when each kind is checked again.
- LLM cost optimization
How to spend less on AI without guessing: pick the cheapest model that does the job, keep chats short, and know the price before you send.
- A hosted MCP client for every model
Add remote MCP servers and use their tools with every llmwise model. Header values are encrypted; tool calls ask for approval by default.
- Our test runs: 50 prompts on every model
The 50 fixed prompts we run on every model in llmwise, how each reply is scored, the grader and its exact prompts, and every result.
- Prompt optimization
Prompt optimization for everyday use: give context, show examples, ask for a format, and get better answers in fewer messages.
- What is a context window?
What a context window is, what counts toward it, what happens when a chat outgrows it, and the context window of every model in llmwise.
What models cost
API prices from our model catalog, dated, and what the same models cost per message in llmwise.
- Claude API pricing
Claude API prices per million tokens as of October 2026, what a typical message costs, and what each Claude model costs per message in llmwise.
- OpenAI API pricing
OpenAI API prices per million tokens as of October 2026, what a typical message costs, and what each GPT model costs per message in llmwise.
- Gemini API pricing
Gemini API prices per million tokens as of October 2026, what a typical message costs, and what each Gemini model costs per message in llmwise.
- DeepSeek API pricing
DeepSeek API prices per million tokens as of October 2026, what a typical message costs, and what each DeepSeek model costs per message in llmwise.
- Grok API pricing
Grok API prices per million tokens as of October 2026, what a typical message costs, and what each Grok model costs per message in llmwise.
- The cheapest LLM APIs, compared
Claude Haiku 5.5 is the cheapest LLM API in llmwise: $0.0008 for a typical message at API prices as of October 2026. Every model, ranked by price.
- LLM cost calculator
What every model in llmwise costs through its API for your tokens and requests a month, and the plan that covers it. Prices as of October 2026.
- What an AI message costs, model by model
What one message costs on every model in llmwise, how far each plan's messages go in each price band, and what the same message would cost at API prices.
Claude, GPT, Gemini, DeepSeek, Grok, Kimi, GLM, and Mistral, in one chat.
See what a message costs before you send it. Free is 5 messages to try; sign in with an email link, no password or card.