Polls

Which FLUX LoRA defines the spring 2026 vibe for you?

We're seeing a wave of new FLUX LoRAs land every week. Which aesthetic actually got into your daily prompts?

What got you into AI generation in the first place?

Pick your AI assistant's default personality

If you could only have one mode for every conversation, which feels least exhausting after a week?

Quiz: how many parameters does FLUX.1 [dev] have?

Quiz: who is the primary developer behind ComfyUI?

Quiz: which company released FLUX.1?

Pop quiz from the model-lineage trivia bin. One right answer.

Most concerning AI capability right now? (pick up to 2)

How should governments regulate AI training in 2026-2027?

Should AI-generated images be watermarked or labelled by default?

Training data: strict opt-in, fair use, or somewhere between?

The single most argued-about question in generative AI right now. Where do you actually land?

Will video generation overtake image generation in popularity by end of 2027?

AGI by 2030 — likely?

Set aside the marketing — your honest read on whether a system meaningfully matches a human across most cognitive tasks within the next ~4 years.

Will open-source catch closed-source by end of 2027?

Image generation quality — have we hit a plateau?

FLUX, Imagen, MidJourney v7 all landed close to each other. Is the field consolidating or just catching its breath?

Which base-model attributes matter most to you? (pick up to 3)

Picking a base checkpoint is a series of trade-offs. Which axes win the trade for you?

Which image architecture do you reach for first in 2026?

Not which is technically best — which one you actually pick when you open ComfyUI today.

Roughly how much do you spend on AI tools per month?

Where do you actually run your image generation?

Multi-GPU for AI generation — game-changer or marketing fluff?

RTX 5090 — worth the $2,000+ price tag for AI workloads?

Real-world minimum VRAM for running FLUX dev locally?

Ignore the marketing slides — what's the smallest card you've actually had FLUX dev render on without GGUF-style quantisation?

Negative prompts in the FLUX era — still earning their keep?

FLUX claims it doesn't need them. SDXL/Pony pipelines still lean on them heavily. Where do you actually stand?

Inpainting — how often does it land in your workflow?

How long do you typically iterate on a single prompt?

From the first version to the version you actually save — how patient are you with one idea?

Daily driver UI for local generation in 2026?

Which interface do you actually open every day? Workflow tools come and go — vote for the one that sticks.

Quiz: which of these base models did FLUX-family LoRAs train on?

Quick one-tap quiz. Pick the base architecture that FLUX-family LoRAs were actually trained against.

Should we add a public ComfyUI workflows directory to UncensoredHub?

Some readers have asked for a one-stop place to share + remix workflow JSONs. We'd start with curated submissions and a stars-style rating.

Which output type are you generating most this month?

Quick pulse-check on what people are actually generating right now — pick all that apply.