Frequently Asked Questions
General / Getting Started
Who is Botscanner for?
Botscanner is beneficial for anyone who uses LLMs and wants optimal results quickly. This includes students, researchers, content creators, marketers, software developers, business analysts, and teams leveraging AI for tasks like writing, coding, research, or analysis.
What is Botscanner?
Botscanner is a platform designed to help you find the best answers from Large Language Models (LLMs) efficiently. It automatically queries multiple AI models (like GPT, Claude, Llama, Gemini, etc.) with your prompt and then uses LLMs themselves to rank the responses based on quality, accuracy, and relevance. This saves you the time and effort of manually comparing different AI outputs.
How do I get started with Botscanner?
Getting started is easy! Just sign up for an account via the "Sign Up" link. As a new user, you'll receive $3 worth of free credits to explore the platform and start comparing model responses immediately.
How Botscanner Works
How does the process work?
It's a simple 4-step process:
1. Enter the question or prompt you want an answer for.
2. Select the 'answering' LLMs you want to generate responses and the 'ranking' LLMs you want to use for evaluation.
3. Botscanner sends your prompt to the selected answering models in parallel.
4. The generated responses are then evaluated and ranked by the selected ranking models. You receive a consolidated list of answers, ordered from best to worst, along with cost and token usage details.
How does Botscanner rank the answers?
Botscanner employs Large Language Models (LLMs) to rank the very same LLMs. This is its magic. You select specific 'ranking' LLMs that evaluate the responses generated by the 'answering' models. These ranking models assess factors like accuracy, relevance, clarity, and adherence to the prompt to score and rank the outputs.
How fast is it?
Generating a task can take from 10 seconds to even more than a minute, depending on the complexity of your prompt and the speed of the models selected for answering and ranking. Tasks are processed asynchronously and in parallel. This means requests to multiple LLMs are sent out concurrently to speed up the overall process. Some models will answer in seconds. More powerful models (especially so called "thinking" models) could take even minutes to answer. But, don't worry. All answers are made available as soon as they arrive and you can already browse through the best ranked answers that are available at the moment. The platform also includes monitoring for stalled tasks and models not responding.
Can I see results before the entire task is finished?
Yes! Botscanner supports displaying partial results. As soon as some models have finished generating and ranking responses, they will appear in your results list, even if other selected models have not finished yet, allowing you to work with initial findings even while others are still processing.
Models & Tiers
Which AI models does Botscanner support?
Botscanner provides access to over 20 models from major AI providers, including OpenAI, Anthropic, Google, Meta, Mistral, DeepSeek, NVIDIA, Alibaba, and xAI. Please check the "Models" page for the most current list and details.
What are Tiers and how do they work?
Tiers control your access level to different models. The "Free" tier provides access to a wide range of capable models. Higher tiers (like Tier 1) unlock access to more advanced, specialized, or resource-intensive models. Your tier automatically upgrades based on your cumulative credit purchase history – the more credits you buy over time, the higher the tier you unlock.
How do I know which tier a model belongs to?
The "Models" page clearly lists the required tier for each available model. If a model requires a tier higher than your current one, it will be indicated, often alongside information about how upgrading works. Your current tier is displayed in your user profile.
Can all models generate answers AND rank responses?
Not always. Some models excel at generating creative or informative text (answering), while others are better suited for evaluating quality (ranking). Many models can perform both roles. The "Models" page specifies the designated role(s) for each model ("answer", "rank", or "both").
Pricing & Credits
How does pricing work on Botscanner?
Botscanner operates on a pay-as-you-go credit system. You purchase credits, and these are deducted based on the specific models you use (both for answering and ranking) and the amount of text processed.
Are there any free credits to start?
Yes, absolutely! Every new user receives $3 in free credits upon signing up, allowing you to test the platform and its features without immediate commitment.
How can I purchase more credits?
You can easily buy more credits through our platform using the secure Stripe Checkout process. Navigate to the 'Add Credits' or 'Billing' section in your account. Your credit balance is updated automatically upon successful payment confirmation (handled via Stripe webhooks).
Account Management
How do I manage my account?
You can sign up, log in, view your profile, check your credit balance, see your current tier, and manage support tickets through the user interface after logging in.
Is my password secure?
Yes. We prioritize your security. Passwords are stored using strong, industry-standard hashing algorithms (specifically, Werkzeug's implementation). We never store passwords in plain text.
Features
What if I need help or encounter an issue?
We offer a built-in Support Ticket system. If you have questions or run into problems, you can create a new ticket, categorize it, describe the issue, and communicate directly with our support team. You can track the status and conversation history of your tickets within the platform.
What are Model Presets?
Model Presets are saved configurations of your favorite answering and ranking models. You can create custom presets (e.g., "My Creative Writing Setup") or use pre-defined ones ("Top Performers", "Affordable", "Fastest"). This allows you to quickly load your preferred model combinations for new tasks with a single click. Presets are saved to your user account.
Is there a guide for new users?
Yes! When you first log in after registering, you'll be offered an interactive tour. This tour provides a step-by-step walkthrough of the platform's main features and how to use them effectively. You can choose to take it then, skip it, or revisit it later from your profile settings.
Troubleshooting & Support
I paid for credits, but my balance hasn't updated. What should I do?
Credit updates rely on confirmation from Stripe via webhooks, which is usually instantaneous. Network delays or rare glitches can occasionally slow this down. If your balance doesn't update within a few minutes, please create a support ticket with details of your purchase (like time or amount, but *never* full card details), and our team will investigate and resolve it. The system includes checks to prevent adding credits twice for the same payment.
A task is taking a very long time or seems stuck. What's happening?
Processing time depends on the selected models' complexity and current load on the external APIs. While we use parallel processing and timeouts, delays can occur. If a task fails or times out, you'll typically see an error message in the results, and you won't be charged for the parts that didn't complete. If you experience persistent slowness or failures, please let us know via a support ticket.
Where can I find answers to other common questions?
You're looking at it! This FAQ page aims to cover the most common queries. If your question isn't answered here, feel free to reach out via the support ticket system.
About
Who is behind Botscanner?
Botscanner is a project of eZecute S.R.L., a leading Italian startup builder, advisor, and investor. eZecute has supported and invested into leading international startups such as Kiwi Bot (now Robots.com), Cricket One, MyFoody, Serket, Game Pix, Evja and many more. eZecute has developed and managed international initiatives such as Startupbootcamp Foodtech, the first agrifood-tech accelerator in Europe, and Hacknight the first Facebook Hackathon in the EU. eZecute is also the founder of AutoBench, the open source LLM benchmarking system based on the Collective-LLM-As-A-Judge methodology. eZecute was founded by Peter Kruger, serial entrepreneur, investor, advisor with a background in theoretical physics of neural networks.
Is Botscanner a startup?
Botscanner is currently an internal project of eZecute but we plan on spinning it out as an independent startup with a fully dedicated team as soon as feasible.
Is Botscanner in beta?
Yes, Botscanner is still at initial stage. While we do our best to ensure the best possible experience to our users, the service may witness occasional restarts or unexpected issues. If you notice anything wrong, please let us know.
What is your vision for the future of AI?
We strongly believe that future major breakthroughs in AI will come from agents that combine the specialized capabilities of different models. Imagine a sort of MoE (Mixture of Experts) but with different models acting as experts. In this vision, evaluating real-time the capability of models to perform a given task, will prove essential. Botscanner is here to help in this development.