AI Resources Hub: ChatGPT and Other Services with Offline Open‑Source Options

A friendly, practical list of AI tools with quick notes on what each one does—and when to consider private/offline options.

General AI Assistants & Research Tools

Google Gemini — chat, multimodal understanding, and Docs/Slides help. Visit
Great for brainstorming, rewriting, and images (in supported regions). Integrates with Google Workspace. Check data policies if using sensitive data.
OpenAI ChatGPT — versatile chat for writing, code, and Q&A. Visit
Excellent general assistant. Use with care for private data; consider redaction or offline alternatives for sensitive content.
Microsoft Copilot — Microsoft’s assistant with Bing search. Visit
Useful for quick answers, image generation, and Office integrations. Enterprise versions offer improved data controls.
Anthropic Claude — strong at long‑context reading and careful reasoning. Visit
Good for summarising large documents and safer drafting. Review T&Cs for data handling if compliance is key.
Perplexity — research assistant with citations. Visit
Ideal for quick research with source links. Helpful to validate facts before acting on them.

Google Gemini (Images) — image generation and understanding. Visit
Create images from prompts and understand images (availability varies by region/account).
OpenAI ChatGPT (Images) — image generation via integrated models. Visit
Generate and iterate on images from natural language prompts alongside chat.
Adobe Firefly — generative images/text effects, integrated with Creative Cloud. Visit
Great for marketing visuals and brand‑safe assets. Check licensing for commercial use.
Midjourney — high‑quality image generation (Discord‑based). Visit
Fantastic artistic output. Useful for concept art and ideation.
Stable Diffusion — open‑source image models (local or cloud). Visit
Flexible for custom styles and private setups. Good option for offline image workflows.
ComfyUI — node‑based workflows for image pipelines. Visit
Visual graph editor to build and automate image generation/fine‑tuning workflows.

ComfyUI (Local) — build private, reproducible image pipelines. Visit
Run locally or on‑prem; great for controlled, auditable workflows.
Ollama — run LLMs locally with simple commands. Visit
Handy for private chat and prototyping on your own machine or server.
LM Studio — desktop app to run local models. Visit
Friendly UI for trying models without coding. Good for demos and small teams.
Stable Diffusion (Local) — image generation on‑prem. Visit
Popular web UI options exist for private image pipelines and fine‑tuning.
FLUX (local models) — modern image model family. Visit
Variants like FLUX.1 can run locally with the right setup; check model cards and licenses.
WAN 2.2 — community image model (niche use cases). Browse
Explore on Hugging Face; review license and requirements before use.

Need help choosing or deploying? We specialise in private & offline AI so sensitive data never leaves your network.

Tool	Best for	Privacy posture	Notes
Gemini	Workspace help Brainstorm	Cloud	Great for docs/slides integration.
ChatGPT	Writing Coding	Cloud	Strong general assistant.
Claude	Long docs Careful reasoning	Cloud	Good for summaries and safer drafting.
Perplexity	Research Citations	Cloud	References sources for validation.
Firefly	Marketing assets	Cloud	Brand‑friendly image generation.
Midjourney	Concepts Art	Cloud	High‑quality art, Discord workflow.
Stable Diffusion	Custom styles	Cloud/Local	Flexible; supports on‑prem setups.
Ollama / LM Studio	Private chat On‑prem	Local	Great for sensitive data & trials.
ComfyUI	Pipelines Automation	Local	Visual, node‑based workflows; great for reproducibility.
FLUX	High‑quality images	Local/Cloud	Modern model family; check variants & licenses.
WAN 2.2	Specialised styles	Local/Cloud	Community model; verify license and requirements.

Prefer a hands‑on session? Check our workshops—we tailor content for beginners to advanced teams.

Never paste sensitive or regulated data into public tools; prefer offline or enterprise options.
Review licensing and usage rights for generated media, especially in commercial settings.
Validate facts with cited sources; treat outputs as drafts, not final truth.

Last updated: September 2025