Help & FAQ

Visit the registration page, enter your email and password to sign up. You'll receive free tokens upon registration and can start using AI chat immediately.

On the login page, click "Forgot Password", enter your registered email, and a verification code will be sent. Use the code to set a new password.

Yes. After logging in, go to your profile settings to change your display name. The email address serves as your account identifier and cannot be changed after registration.

GPTGet integrates multiple leading LLMs including GPT-4o, GPT-4.5, Claude 3.7 Sonnet, Gemini 2.5 Pro, DeepSeek, and more. Available models vary by subscription plan. See our pricing page for details.

A token is the basic unit for AI model text processing. Roughly 1 Chinese character = 1-2 tokens, 1 English word = 1 token. Each conversation consumes tokens for both input and output. You can view detailed token usage on your usage page.

Click the attachment button (paperclip icon) next to the chat input, then select the file to upload. Supports images, PDFs, text files and more. Uploaded files will be analyzed by AI, and you can ask questions about the content.

Yes. GPTGet supports web search. The AI can search the internet in real-time to get the latest information for your queries. Simply ask in the chat, and the AI will automatically determine whether a search is needed.

The memory feature allows AI to remember your preferences and past conversation context. AI automatically extracts your personal info and preferences (like language, expertise areas, etc.) to provide more personalized responses. You can view and manage these memories on the memory management page.

GPTGet offers a free tier and multiple paid plans. The free tier includes tokens on sign-up for basic features. Paid plans offer more token quotas and access to premium models. See our pricing page for details.

After logging in, visit the "Subscription" page, choose a plan that suits you, click "Subscribe", and follow the payment instructions. Your token quota will be activated immediately after successful subscription.

Currently we support Alipay and cryptocurrency (USDT) payments. We are continuously adding more payment methods.

With auto-renewal enabled, the system will automatically renew and charge when your subscription expires. You can disable auto-renewal anytime on the subscription management page; your current subscription remains active until expiry.

When tokens run out, you won't be able to start new conversations. You can: upgrade to a higher subscription plan, purchase a token top-up package for extra quota, or wait for your monthly quota to reset.

Visit the paper submission page, paste an Arxiv paper URL or ID (e.g., 2301.12345), and click submit. The system will automatically fetch the paper info and generate a structured AI analysis.

Each paper analysis includes: translated title, TL;DR one-line summary, research background & motivation, core methodology, key findings & contributions, experimental results, limitations & future directions, plain-language explanation, glossary, and related code repository links.

AI analysis is typically completed within 1-3 minutes after submission. Complex papers may take longer. Once ready, you can view the full analysis on the paper detail page.

Upload a PDF file in chat, or open it via the "PDF Reader" button on a paper detail page. The reader displays the original PDF with AI-assisted features.

The PDF reader supports: AI summary generation, smart table of contents extraction, paragraph translation, term explanation, key passage highlighting, keyword extraction, reference parsing, AI Q&A (based on document content), and more. Select text to directly invoke AI explanation or translation.

A workflow is a visual AI task orchestration tool. You can chain multiple AI operations (text generation, data processing, conditional logic, etc.) into automated pipelines for batch processing of complex tasks. Examples: batch document translation, automated report generation, data analysis pipelines, etc.

Available node types include: LLM node (call AI models), variable node (set/transform data), conditional branch node, loop node, HTTP request node, code execution node, text template node, human review node, and more. Additional node types are continuously being developed.