lora_tester

Generate images and videos via Comfy Cloud (cloud.comfy.org) instead of a local server. Use when the user wants to run ComfyUI workflows on cloud infrastructure, or when local GPU isn't available.

Generate images and videos via the local ComfyUI server on port 8188. Use when the user wants to generate, create, or render images/video using a locally-running ComfyUI installation.

Shared best practices and proven workflow recipes for ComfyUI (Z-Image Turbo, LTX 2.3, Wan 2.2, etc.). Backend-agnostic — workflow JSON is identical between local and cloud. Loaded by /comfy_local and /comfy_cloud; can also be used standalone for workflow-building reference.

Systematically test LoRAs across multiple prompts and strength values to compare them visually. Creates self-contained project folders with a manifest and downloaded images, viewable in a single-file HTML gallery (grid, side-by-side, A/B slider, PNG export). Use when the user wants to test, compare, or evaluate a LoRA. Two backends: lora_test.py (local ComfyUI) and lora_test_cloud.py (Comfy Cloud).