Which base model should I use with the Oldschool Hud Graphics LoRA?

Use a checkpoint that matches the LoRA’s architecture (e.g., SD 1.5 LoRA with an SD 1.5 checkpoint, SDXL LoRA with an SDXL checkpoint). Check the LoRA’s model card or filename for its target version. Mixing versions will reduce quality or fail to load.

How do I control how strong the HUD style appears?

Adjust the LoRA weight (lower for subtle, higher for strong). Refine prompts to emphasize geometry (reticle, grid, ring, UI readouts) and use a negative prompt to suppress photoreal features. If outputs feel busy, slightly lower CFG or reduce steps to simplify shapes.

My lines look soft or messy—how can I sharpen them?

Increase sampling steps modestly in Z‑Image Turbo, keep CFG moderate, and try a resolution that matches your final layout (avoid extreme upscales in one pass). Tighten the prompt to fewer, clearer elements (e.g., one reticle + one grid) and avoid conflicting style terms.

How do I turn a batch of images into a video?

After generating a batch with BatchImagesNode, connect the sequence to CreateVideo, set your frame rate and ordering, then export with SaveVideo. This compiles your HUD variants into a short motion clip for previews or social posts.

Z-Image-Turbo_Oldschool-Hud-Graphics

This workflow generates retro, old‑school HUD graphics from text prompts by combining a style LoRA with a turbo sampler for rapid iteration. The "Oldschool Hud Graphics" LoRA steers the base checkpoint toward geometric UI motifs—reticles, gridlines, rings, readouts—while the Z‑Image Turbo node accelerates sampling for fast previews and quick style exploration. BatchImagesNode lets you spin up variations in one pass, and SaveImage captures your chosen frames. When you’re ready to animate, CreateVideo and SaveVideo assemble your batch into a short motion clip.

Under the hood, your prompt is encoded and fed to the base model with the LoRA applied at a user‑set weight, biasing attention layers toward the HUD aesthetic. Z‑Image Turbo then samples with low step counts for speed, which you can tune for crisper lines or more detail. Keep the LoRA weight moderate to strong to emphasize the graphic look, and pair it with concise prompts (e.g., "retro heads‑up display, concentric circles, grid overlay, monochrome green") plus a negative prompt to suppress photoreal clutter. The included MarkdownNote contains in‑graph tips, while the custom nodes (listed by UUID) handle the LoRA/Turbo plumbing so you can focus on prompts, batch size (BatchImagesNode), and output (SaveImage/SaveVideo).

API

Use this workflow from code

Every Comfy workflow is a JSON graph. The payload below is this workflow, exactly as ComfyUI runs it — fetch it from the URL, keep it in version control, or load it in ComfyUI and run it node by node.

36794f399aa0.json

Fetching workflow JSON…

Run it from TypeScript or Python with the Comfy SDK. The same code targets Comfy Cloud or a ComfyUI you host yourself — only the base URL changes.

// Install (beta)
npm i @comfyorg/sdk

// Run this workflow (TypeScript)
import { Comfy } from "@comfyorg/sdk";

const client = new Comfy({ apiKey: "comfyui-..." });
const wf = await client.workflows.fromFile("workflow_api.json");
const job = await client.run(wf);
await job.getOutputs("<output-node-id>")[0].toFile("output.png");

The SDK takes a workflow in API format: open this workflow in ComfyUI and use File → Export Workflow (API).

Comfy Cloud API access requires a plan with an API key.

SDK docs Get an API key

الأسئلة الشائعة