AI creation toolkit
How to use
DeepAI.
Quickly exploring AI chat and visual concepts, generating or editing images, producing short visual drafts, and integrating straightforward media endpoints into prototypes.
DeepAI is a Consumer and developer AI toolkit for chat, image and video generation, image editing, upscaling, colorization, and background removal. Quickly exploring AI chat and visual concepts, generating or editing images, producing short visual drafts, and integrating straightforward media endpoints into prototypes. This guide covers the whole path in one place: official access, a first session that produces something reviewable, the checks that make output trustworthy, and the permissions worth limiting before you connect real work.
AI should make the work easier to inspect. If the workflow removes the source, the owner, or the review step, redesign the workflow.
Access & setup
Find, install, and sign in to DeepAI
Get into the official DeepAI experience with the right account and a setup you understand.
- 01
Start at https://deepai.org/ and confirm the domain before entering account or payment information.
- 02
Availability: DeepAI is available through its website and simple HTTP APIs for selected generation and image-processing tools.
- 03
A modern browser for interactive tools or a Pro account and secret API key for API access, plus rights to all prompts and uploaded assets and human review before use.
- 04
Sign in with the account you intend to keep using, then review plan, data, notification, and permission settings.
- 05
Run one low-risk test task before connecting sensitive files, repositories, or workspace data.
- Use official download pages.
- Review permissions during setup.
- Keep installers and applications updated.
First session
Your first useful DeepAI session
Learn the interaction loop using a small task with a clear outcome.
- 01
Define the purpose, audience, medium, subject, style, rights, prohibited content, disclosure, and concrete acceptance criteria before prompting.
- 02
Use a minimal non-sensitive prompt or authorized image and select the narrowest tool and lowest useful resolution for the first test.
- 03
Generate a small set and inspect accuracy, anatomy, text, identity, bias, artifacts, provenance, and whether the result makes an unsupported factual or commercial claim.
- 04
Revise deliberately, finish exact copy externally, verify current plan and terms, obtain human approval, and preserve the prompt, source, result, and disclosure record.
- State the outcome before the background.
- Provide the real source material.
- Review the result before expanding the task.
Quality control
Check the quality of DeepAI output
Turn generated code into a change you would put your name on in review.
- 01
Restate the intended behaviour and the acceptance criteria before reading a single generated line.
- 02
Read the diff rather than the chat summary; the summary describes intent, the diff describes what actually changed.
- 03
Run the test suite, then write a test that would fail if the change were wrong, and confirm it passes for the right reason.
- 04
Check the edges the model tends to skip: error paths, null and empty cases, concurrency, and migration or rollback behaviour.
- 05
Trace every new dependency, network call, and file write, and confirm each one is genuinely required.
- Review the diff, never the description of the diff.
- A passing test suite proves the tests pass, not that the change is correct.
- Keep changes small enough that a human can actually review them.
Privacy & permissions
Use DeepAI safely
Give the tool enough repository access to help, and no more than that.
- 01
Decide what the tool may read and what it may write before connecting a repository.
- 02
Scope tokens and integrations to the narrowest repository, branch, and permission set that still works.
- 03
Keep secrets out of the context window: use environment variables and secret stores, never inline credentials.
- 04
Require review before generated code touches authentication, payments, permissions, migrations, or deletion paths.
- 05
Audit what the integration retained after the session, and revoke access that is no longer needed.
- DeepAI says users own their input and output but may use User Content to provide, develop, improve, secure, and enforce its services. Generated content can be inaccurate, derivative, unsafe, or visually misleading, and ownership language does not clear third-party copyright, trademark, likeness, or privacy rights. Never expose API keys, upload confidential material without approval, or automate publication without review.
- Follow your organisation's approved-use policy.
- Never treat fluent output as authorization to act.
Core workflows
Step-by-step ways to use DeepAI for the work it does best.
Each workflow below has its own outcome, steps and review checkpoints.
Turn a written brief into an image draft while protecting rights, identity, factual meaning, and the human approval boundary.
↓ 02 WorkflowIntegrate a DeepAI image endpoint safelyAdd generation or image editing to a prototype with secret handling, bounded inputs, cost controls, and a mandatory review queue.
↓Workflow 01
Develop a reviewable visual concept in DeepAI
Turn a written brief into an image draft while protecting rights, identity, factual meaning, and the human approval boundary.
Outcome: Produce a useful concept image whose source, prompt, limitations, and approval status remain clear.
- 01
Write a brief with audience, placement, subject, composition, style, exact copy space, prohibited elements, rights, disclosure, and acceptance criteria.
- 02
Use a non-sensitive prompt and the lowest useful resolution, avoiding living-artist imitation, unauthorized people, protected characters, and unverifiable factual scenes.
- 03
Generate a small set, record the tool and settings, and choose on composition rather than polishing the first plausible output.
- 04
Inspect anatomy, identity, logos, text, cultural representation, background logic, artifacts, source similarity, and any factual or commercial implication.
- 05
Revise deliberately, create exact typography externally, obtain brand and rights approval, disclose material generation, and archive the brief, prompt, candidates, and final.
What good use looks like
- Start with a written acceptance test.
- Keep exact text outside generation.
- Never imply an unverified real event.
Workflow 02
Integrate a DeepAI image endpoint safely
Add generation or image editing to a prototype with secret handling, bounded inputs, cost controls, and a mandatory review queue.
Outcome: Run a predictable media API that cannot expose its key or publish unsafe output automatically.
- 01
Confirm the current endpoint, plan, parameters, size limits, terms, User Content handling, moderation, retention, and allowed product use.
- 02
Keep the API key server-side, authenticate owners, validate file type and dimensions, remove metadata, and require rights and sensitivity attestations.
- 03
Submit idempotent jobs with bounded prompts, resolution, timeout, retry count, concurrency, and per-user cost limits while preserving the untouched input.
- 04
Quarantine results for malware, format, dimensions, visual safety, identity, text, artifacts, rights, and task-specific human quality review.
- 05
Release only through an approved action, redact logs, rotate keys, reconcile spend, test provider errors and duplicates, and delete temporary assets on schedule.
What good use looks like
- Keep keys off the client.
- Limit size, retries, and spend.
- Quarantine every generated asset.
Official references
Check the current product documentation.
Features, plan limits, availability, and data controls change. These official pages are the starting points used for this guide.