DeepAI

What is DeepAI?

DeepAI is a multimodal creation platform operated by Deep AI, Inc. in the United States. It started as a tool for generating images in browsers, and today it offers capabilities for chatting, creating images, videos, music, handling voice inputs, as well as providing developer interfaces. It can be used directly in web browsers, and there are also iOS and Android apps as well as Chrome extensions available for it.

Free users can quickly try out the basic tools, while DeepAI Pro offers higher-quality modes, monthly quotas, private generation options, ad removal, and an enhanced wallet feature. This platform is suitable for creation and prototype testing; however, the quotas and pricing vary significantly among the different modes, so it is important to determine the desired quality level before generating content.

Main functions

• AI Chat: Handles tasks such as answering questions, writing, summarizing, translating, as well as working with code and attachments; the Online Mode allows access to the internet, while Genius and Super Genius are designed for more complex tasks involving knowledge, mathematics, logic, and programming.
• Multi-model chatting: Through a unified interface, you can choose either DeepAI’s own models or more than 20 third-party chatting models, which makes it possible to compare the style of responses and the reasoning capabilities of these models; the list of available models is subject to change.
• Image generation: Enter a text prompt to create images in various styles, with three quality and cost levels available – HD, Genius, and Super Genius 2K.
• Photo editing: It allows for modifying images based on text, replacing certain elements within an image, removing the background, coloring black-and-white photos, improving resolution, or reducing noise and enlarging anime images.
• Short video creation: Create short videos with audio from text or images; Pro offers HD as well as the more expensive Hollywood mode.
• Music and voice: Generate original music and engage in real-time voice conversations with AI via Voice Chat.
• Concierge: Multi-step requests are delegated to task-based assistants; a fixed number of such tasks are included per month, with additional tasks charged separately.

Input and output

Basic usage process

1. Open the desired web tool directly for a free trial, or register an account to save history, sync devices, and use the online mode.
2. Select Chat, Image, Editing, Video, Music, or Voice based on the task, and determine whether the current mode is Free, HD, Genius, or a higher quality level.
3. Enter clear objectives, subjects, style, format, and constraints; for editing tasks, it is also necessary to upload the original files for which legitimate usage rights exist.
4. After generation, check the facts, code, details of characters, text, copyright elements, and narrative continuity; adjust the prompts if necessary and generate again.
5. Check the remaining quota before downloading or sharing; for the Pro version, any excess beyond the monthly quota will be deducted from the prepaid wallet, and an acceptable spending limit must also be set when automatic top-up is enabled.

Free version and Pro package

The Pro quota is updated on a monthly basis starting from the date of subscription, rather than on the 1st of each month. It can still be used until the end of the paid cycle after cancellation; the current page does not specify a fixed deadline for refunds or any guarantee regarding refunds.

The Pro plan includes a monthly quota.

Excess wallet and automatic top-up

The wallet can be topped up manually, or automatic top-up can be enabled when the balance runs low. Automatic top-up helps to prevent tasks from being interrupted, but it may also result in additional costs due to repeated attempts or the creation of high-cost videos; it is therefore recommended to check the bills regularly and turn off any unnecessary automatic top-ups.

Developer API

• DeepAI Pro includes a REST image API; authentication is carried out using the account API Key in the request headers, and the response provides the address of the result along with task details in JSON format.
• The current document covers text-to-image generation, image editing and partial redrawing, background removal, black-and-white coloring, super-resolution, Waifu2x, Creative Upscale, and image replacement.
• The interface utilizes standard network requests and JSON, and can be called from Python, JavaScript, the command line, or other languages; no dedicated SDK needs to be installed.
• API calls are counted toward the Pro monthly quota; any excess usage will result in charges from the prepaid wallet. The server must store keys securely, and these keys should not be embedded directly in public web pages or mobile applications.

API integration process

1. Register for and subscribe to DeepAI Pro, then obtain the API Key from your account dashboard.
2. Select the appropriate image endpoint based on the task, prepare the text field or image file, and check the supported sizes and formats.
3. Send a request from your own backend, include the API Key in the request headers, and submit the parameters via a form or an interface.
4. Parse the JSON response; handle timeouts, failures, content filtering, and duplicate requests before saving the output file, thereby preventing uncontrolled retries from depleting the wallet balance.
5. In the production environment, record the number of calls made, the cost per call, and the error rate; regularly rotate keys and restrict the services that have access to them.

Platform and expansion

Privacy and data usage

• The platform collects names, email addresses, phone numbers, payment information, and communications with customer service, as well as user-generated content such as prompts, files, images, text, videos, and audio.
• It also automatically collects information on devices, systems, network addresses, browsers, times of access, pages, functions, clicks, and third-party login details, and may be able to determine the approximate location.
• The information is used for service delivery, product improvement, marketing, targeted advertising, analysis, security, and training of DeepAI’s artificial intelligence models.
• The privacy policy permits the disclosure of data to suppliers, advertising partners, consultants, law enforcement agencies, and affiliated parties; it may also allow certain information to be shared with third parties for commercial purposes or for training models.
• Users can opt out of model training, partial sale or sharing, as well as targeted advertising through privacy settings; when not logged in, this setting may be tied only to the current browser, and it needs to be reset when switching devices or deleting cookies.
• An account can be deleted from the dashboard or a deletion request can be submitted; once processed, the account data is usually removed within 7 days, although any email exchanges may remain intact.

The “private generation” option in Pro relates to the visibility of the created content; it cannot be automatically interpreted as meaning that the content will not be used for training or disclosed to third parties. Before processing customer files, unpublished materials, or personal data, it is necessary to set up privacy-related settings separately and assess whether it is appropriate to upload such data.

Copyright, Commercial Use, and Usage Restrictions

• The DeepAI terms state that it does not claim copyright over the content generated by its tools and APIs, and allow for commercial use within legal limits.
• This type of platform license does not guarantee that the output will be protected by copyright in the country where it is used, nor does it ensure that it will not resemble third-party works, trademarks, portraits, or materials.
• Users must not engage in illegal activities, harm minors, create malicious false information, harass or discriminate against others; nor shall they use the service to provide medical advice or make fully automated decisions that affect others’ legal rights.
• Users who are under 18 years of age or have not reached the local legal age of majority may use it only under the supervision of a parent or guardian.

Open-source status

The DeepAI web platform, hosted models, and commercial APIs are not available as fully open-source products. The older JavaScript and C# clients available on the official GitHub repository have been archived and are licensed under BSD-2-Clause; there are also separate projects licensed under Apache-2.0, MIT, and other licenses. The license associated with each repository applies only to that repository itself and does not extend to the platform or the models generated.

Suitable for users and their capacity limits

• Regular creators: Use a single account to work with images, short videos, music, and chat, without the need to subscribe to multiple separate tools.
• Design and content team: Quickly create visual drafts, remove backgrounds, apply colors, zoom in, and produce short video clips.
• Developer: Image generation and editing can be integrated using simple REST interfaces, but it is necessary to set up one’s own keys, security measures, queues, caching systems, and mechanisms for cost control.
• Researchers and writers: Online chat and multiple model options can be used to assist with searching and comparison, but the responses must be verified; the online mode does not guarantee that the information is complete or accurate.
• Long-duration videos, high-quality film and television production, highly consistent character designs, and the processing of regulated data are not among the main advantages of its low-cost packages; these tasks require additional tools and manual procedures.

Frequently Asked Questions

Can DeepAI be used for free?

A free trial is available, offering limited standard chat, HD image editing, music, and 60 seconds of voice communication per day. The free version contains ads and does not include videos, the Genius mode, memory functions, online browsing, nor an extra budget.

Is the Pro quota refreshed on the 1st of each month?

No, the quota is based on the date of initial subscription and is updated every month. For annual subscriptions as well, the usage quota is provided on a monthly basis, rather than being granted in one lump sum for the entire year.

What happens when the limit is exceeded?

Additional usage by Pro users is deducted from the prepaid wallet at the unit price per item, and the wallet must maintain a positive balance. Once automatic top-up is enabled, payment will be made again if the balance becomes insufficient; therefore, it is important to keep an eye on successive attempts to make payments in case they fail.

Can the generated content be used for commercial purposes?

The platform’s terms permit the creation of content for legitimate commercial purposes, and it states that no copyrights are claimed on such content. Users still need to check the relevant rights, ensure there is no similarity with third-party works, consider issues related to portraits, and abide by the regulations governing AI-generated content in their respective jurisdictions.

Will DeepAI use the uploaded content to train models?

The privacy policy permits the use of information for training models, and it may also allow certain user content to be shared with third parties. Users need to actively opt out through the privacy settings, and it should be noted that options related to not being logged in may apply only to the current browser.

Is DeepAI open source?

The core platform and hosting models are not fully open-source products. The vendor maintains several code repositories under different licensing agreements, but the older JavaScript and C# clients have been archived, and those licenses do not apply to the online services.

Summary

DeepAI’s advantage lies in the fact that it offers chat, image, short-video, music, voice, and image APIs under a single account at low subscription costs, along with clear pricing for any additional usage. When using this service, it is important to pay attention to factors such as quality levels, automatic recharging, video duration, privacy settings, and third-party rights, rather than simply focusing on the fact that all tools are included.