Askii FAQ Guide

Askii User Hub
,
1

Getting Started

  1. What is Askii.ai?

Askii.ai is a web-based multi-LLM platform that allows you to interact with multiple AI models simultaneously (GPT-4, Claude, Gemini, etc.), compare responses side-by-side, and manage costs through transparent micro-payments via the Moneta system.

  1. What features does Askii.ai offer?

Askii offers several unique advantages:

  • Multi-model access: Query multiple AI models simultaneously and compare responses

  • Response merging: Intelligently combine responses from different models

  • Transparent micropayments: Pay only for what you use with real-time cost tracking

  • Workspace tools: Create custom AI assistants, knowledge bases, and reusable prompts

  • Usage analytics: Comprehensive tracking and optimization insights

  1. What do I need to get started?

You need to:

  • Navigate to [url to page to discuss registration]

Query Features

  1. How do I do a simple query?

  • Choose an LLM model from the drop down list at the top of the page.

  • Enter your query in the prompt window, and hit return.

  1. How do I do a Multi-model query?

  • Choose an LLM model from the drop down list at the top of the page.

  • Click on the ‘+’ button, to add another Model drop down, choose another LLM model

  • Repeat above to add as many models as you want.

  • Enter your query in the prompt window, and hit return.

  • The results are displayed side by side.

  1. How do I generate an Image?

  • Choose the Image button in the Prompt box.

  • Enter your query in the prompt window, and hit return.

Multi-Model Features

  1. How many AI models can I use simultaneously?

You can select as many models as you like simultaneously for any query. Popular combinations include mixing premium models (Gemini Pro) with cost-effective options (GPT 4o-mini) for comparison of quality of results and for budget optimization.

  1. How does multi-model querying work?

When you submit a query, Askii:

  • Sends your prompt to all selected models concurrently

  • Displays responses side-by-side for easy comparison

  • Shows real-time cost tracking and token usage for each model

  • Allows you to rate and compare response quality

  • Offers optional AI-powered response merging

  1. What is response merging and how does it work?

Response merging intelligently combines the best elements from multiple AI responses. The merging is done using the last LLM chosen.

Workspace Features

  1. What can I do in the Workspace?

The Workspace has three main components:

  • Models: Create custom AI assistants with specialized prompts and knowledge bases

  • Knowledge: Upload documents and create information sources for enhanced context

  • Prompts: Build reusable prompt templates with variables

  1. How do I create a custom AI assistant?

To create a custom model:

  • Go to Workspace → Models → Click ‘+’

  • Enter a Model Name (e.g., “HR Recruitment Assistant”)

  • Add a System Prompt for specialized behavior

  • Attach relevant Knowledge Bases

  • Choose your preferred LLM

  • Save the model for future use

  1. What are Knowledge Bases and how do I use them?

Knowledge Bases are collections of documents (PDFs, text files, web content) that enhance AI responses with your specific information. Create them in Workspace → Knowledge, then access them by typing ‘#’ in the chat and selecting your knowledge base.

  1. How do Prompt Templates work?

Prompt templates let you create reusable prompts with variables. For example:

  • Template: “Create a {{PLATFORM}} post for {{INDUSTRY}} targeting {{AUDIENCE}}”

  • Access by typing ‘/’ in chat

  • Fill in variables when prompted

  • Reuse for consistent formatting

Pricing and Cost Management

  1. How does Moneta micropayment work?

Moneta charges you based on actual usage rather than flat monthly fees. You see:

  • Exact token usage and pricing for each model

  • Detailed cost breakdowns in your usage ledger

  • No subscription fees - pay only for what you use

  1. What are tokens and how am I charged?

Tokens are the units used to measure AI usage. Think of tokens as pieces of words - roughly 4 characters or 3/4 of a word in English. You are charged for:

  • Input tokens: The number of words/characters in your query, including any uploaded documents or knowledge base content

  • Output tokens: The number of words/characters in the AI’s response

  • Different AI models have different pricing per token

  • You can see exact token counts and costs in real-time before submitting querie

  1. How can I track and optimize my spending?

Use the Usage Dashboard to view:

  • Daily/weekly/monthly spending trends

  • Cost per query by model

  • Token efficiency metrics

  • Optimization recommendations (like switching to cheaper models for simple tasks)

  1. Is Askii prepaid or postpaid?

[This section will be updated with specific payment structure details - whether users need to load credits in advance or are billed after usage]

  1. How can I control my costs?

  2. What if I have a very low budget but need complex analysis?

Several strategies can help maximize value on a tight budget:

  1. Strategic model selection: Use cost-effective models like Gemini Pro for initial analysis, then upgrade to premium models only for final refinement

  2. Staged querying: Break complex requests into smaller, focused queries to better control token usage

  3. Knowledge base optimization: Upload relevant documents once to reduce context needed in each individual query

  4. Prompt templates: Create efficient, reusable prompts to minimize token waste and ensure consistent results

  5. [Budget assistance programs and minimum spend options to be detailed]

  6. Do Askii accounts expire?

  7. Are there any minimum spending requirements?

  8. What happens if I run out of credit mid-query?

  9. Can I set spending limits or budgets?

  10. Is there a free trial or starter credit for new users?

Advanced Features

  1. Can I upload documents for analysis?

Yes, you can:

  • Upload documents directly in the chat for one-time analysis

  • Create permanent Knowledge Bases for repeated use

  • Access multiple knowledge bases simultaneously using ‘#’ in chat

  • Support for PDFs, text files, and other document formats

  1. How do I access my chat history?

Your chat history is available in the left sidebar. Each conversation includes:

  • Full query and response history

  • Models used and costs incurred

  • Workspace resources utilized

  • Timestamps and usage metrics

Technical Questions

  1. Is my data secure and private?

Yes, Askii provides privacy and control features including:

  • Option to deploy specialized AI assistants locally

  • Controlled environments for sensitive data

  • Comprehensive usage logging without compromising privacy

  • Transparent data handling policies

  1. What happens to my custom models and knowledge bases?

Your workspace resources are:

  • Saved permanently to your account

  • Accessible across all your chats

  • Searchable using universal search (/, #, @)

  • Transferable and exportable (contact support for details)

Troubleshooting & Support

  1. What if I need help or have technical issues?

For support:

  • Check the comprehensive user documentation

  • Use the built-in help features in the interface

  • Contact our support team through the platform

  • Access troubleshooting guides in your account dashboard

Quick Reference

Access Workspace Resources:

  • Type @ for custom models

  • Type # for knowledge bases

  • Type / for prompt templates

Multi-Model Tips:

  • Mix premium and budget models for optimization

  • Use response merging for comprehensive analysis

  • Check real-time costs before submitting queries

Cost Optimization:

  • Use Gemini Pro for simple tasks (lowest cost)

  • Reserve GPT-4 for complex analysis

  • Create prompt templates for repeated queries

  • Monitor usage dashboard for optimization suggestions

This FAQ is regularly updated based on user feedback and new features.