git:20260626.80f5c48 to git:20260626.f793262

52 added, 25 removed. Audit A to A.

---

name: krea2-txt2img

description: Build Krea 2 Turbo txt2img workflows — native krea2 CLIPLoader, Qwen3-VL encoder, Qwen image VAE, 8-step turbo settings, and Ideogram-style JSON prompting

globs:

- "**/*.json"

---

# Krea 2 Text-to-Image Workflows

## Overview

Krea 2 is a **12B-parameter Diffusion Transformer** from Krea.ai (released June

2026, weights open-sourced under the Krea 2 Community License — free commercial

use up to 50 seats). Two variants:

1. **Krea 2 Raw** — the base checkpoint before extra post-training. For

fine-tuning / maximum fidelity, more steps.

2. **Krea 2 Turbo** — post-trained + **distilled**; generates in **~8 steps at

- cfg 1**. This is what the `krea2-txt2img` pack ships.

+ cfg 1**. This is what the krea2 txt2img packs ship.

+ ## Two packs (no group toggles, no bypassed nodes)

+ The old single `krea2-txt2img` graph (prompt mode toggled by bypassing nodes) is

+ split into two standalone packs — pick by how you prompt:

+ - **`krea2-txt2img-manual`** — plain prose prompt (the `MANUAL PROMPT` node).

+ - **`krea2-txt2img-json`** — Ideogram-4-style structured JSON / area prompting

+ (`Ideogram4PromptBuilderKJ`).

+ Both are single-pipeline graphs with no group toggles — each pack's one prompt

+ source is active and the other prompt node is removed (no prompt-mode bypass to

+ flip). `ImageSharpenKJ` runs before `SaveImage`. Two **optional** post-proc nodes

+ ship **bypassed** in both (see below). Both are render-verified.

Krea 2 has **native ComfyUI support** (`comfy/text_encoders/krea2.py`, ComfyUI ≥

v0.26.0): the `CLIPLoader` uses **`type=krea2`**, with a **Qwen3-VL 4B** text

encoder and the **Qwen image VAE**. The Qwen3-VL encoder drives strong prompt

adherence and structured-JSON prompts.

## Models (all from the `Aitrepreneur/FLX` mirror; official: `krea/Krea-2-Turbo`)

| Slot | File | Notes |

|---|---|---|

| `diffusion_models/` | `krea2_turbo_fp8.safetensors` | 12B Turbo, fp8 — RTX 4000/3000/2000 |

| `diffusion_models/` | `krea2_turbo_mxfp8.safetensors` | RTX 5000 (Blackwell) native fp8 |

| `text_encoders/` | `qwen3vl_4b_fp8_scaled.safetensors` | Qwen3-VL 4B encoder |

| `vae/` | `qwen_image_vae.safetensors` | Qwen image VAE |

## Node stack

- **core**: `UNETLoader` (krea2_turbo) → `CLIPLoader` (type=krea2) → `VAELoader`

(qwen_image_vae), wired via KJNodes `SetNode`/`GetNode` buses into a subgraph

- (`CLIPTextEncode` → `KSampler` → `VAEDecode`, with an rgthree `Any Switch` that

- picks the manual prompt vs the JSON-builder prompt).

+ (`CLIPTextEncode` → `KSampler` → `VAEDecode`). An rgthree `Any Switch` sits in

+ front of the encoder; in each pack only that pack's prompt source is wired to it

+ (manual node in `-manual`, JSON builder in `-json`).

- **rgthree-comfy**: Power Lora Loader, Any Switch, Label, Fast Groups.

- **ComfyUI-KJNodes**: Set/Get, `Ideogram4PromptBuilderKJ`, `ImageSharpenKJ`.

- - **ComfyUI-RBG-SmartSeedVariance**: `RBG_Smart_Seed_Variance` (opt-in).

- - **ComfyUI-ConditioningKrea2Rebalance**: `ConditioningKrea2Rebalance` — the Krea 2

- **safety-filter bypass**. Ship it **bypassed**; the filter is intact by default.

+ - **ComfyUI-RBG-SmartSeedVariance**: `RBG_Smart_Seed_Variance` — **optional**, ships

+ **bypassed** in the positive-conditioning loop.

+ - **ComfyUI-ConditioningKrea2Rebalance**: `ConditioningKrea2Rebalance` — **optional**,

+ ships **bypassed** (after seed-variance), in the same loop.

## Settings that matter

- **steps 8, cfg 1** — Turbo is distilled; more steps/higher cfg over-cooks it.

- **sampler `er_sde`, scheduler `simple`** — the verified defaults.

- **1920×1080** default; Krea 2 handles a wide aspect range.

- - The JSON prompt-builder (`Ideogram4PromptBuilderKJ`), seed-variance, and

- laplacian-sharpen post-proc ship **bypassed** — opt-in.

+ - The prompt source is fixed per pack (manual node vs JSON builder) — no

+ prompt-mode bypass to flip.

+ ## Optional post-proc (ship bypassed — un-bypass to use)

+ Both packs leave these two nodes in the positive-conditioning loop, **bypassed**

+ (passthrough). Un-bypass on the live canvas with `panel_set_node_mode` (or in the

+ UI) when you want them:

+ - **`RBG_Smart_Seed_Variance`** — controlled variations of the same prompt without

+ changing the composition. Enable it, set its seed mode to `randomize`, and tune

+ the variance mode (e.g. `🌿 Balanced`) / strength widgets. Leave bypassed for a

+ deterministic single result.

+ - **`ConditioningKrea2Rebalance`** — rebalances the per-token conditioning weights

+ (the comma-separated weight string, e.g. `1.0,…,2.5,5.0,1.1,4.0,1.0`). The

+ upstream author frames it as removing Krea 2's built-in **safety filter**, so

+ **leave it bypassed for the model's default safety behavior**; only enable (and

+ adjust the weights) with a specific, authorized reason.

## JSON / area prompting

Like Ideogram 4, Krea 2's Qwen3-VL encoder reads structured prompts (per-region

- desc + bounding boxes + palettes). The pack ships with BOTH a manual-prompt node

- and the JSON builder, picked by an rgthree `Any Switch` — and the JSON builder

- ships **bypassed**, so by default the manual prompt wins. To drive it from the

- JSON builder: **set `Ideogram4PromptBuilderKJ` (node 14) to mode 'active' and the

- manual prompt (node 143) to mode 'bypass'** — on the live canvas use

- `panel_set_node_mode` for both, then re-read with `panel_get_graph` to confirm the

- modes actually flipped (the `Any Switch` feeds whichever is active into the

- encoder). Do NOT assume the switch is already on the path you want — a stale

- bypass here silently renders the wrong (old manual) prompt. After the render,

- VERIFY the image matches the JSON you set (view it) BEFORE continuing; if it

- doesn't, the builder is probably still bypassed or a field is stale — fix and

- rerun. Gotchas learned the hard way:

+ desc + bounding boxes + palettes). For structured prompting just use the

+ **`krea2-txt2img-json`** pack — its `Ideogram4PromptBuilderKJ` drives the encoder

+ directly (no bypass to flip). After the render, VERIFY the image matches the JSON

+ you set (view it) BEFORE continuing; if it doesn't, a field is probably stale —

+ fix and rerun. Gotchas learned the hard way:

- **Set ALL the builder fields**, not just the prompt/boxes — `background`,

`technical`, `style`, `lighting` (widgets 3/5/6/7). Leaving stale values leaks

content (a leftover celebrity portrait bled into a tea still-life).

- **Keep palettes minimal or empty.** A rich top-level palette can render as a

literal color-swatch strip down the edge of the image. Empty `palette: []`

(top-level and per-box) gives a clean full-frame result.

- Add "no people / single full-frame photograph" to `style` for object/landscape

scenes — Krea 2 follows it well.

## Render-verified

- 5 images render crisp at 1920×1080 / 8 steps / cfg 1 / er_sde — 3 prose (wildlife,

- urban night, still-life) + 2 Ideogram-JSON (still-life, landscape). **Note:** the

- `ImageSharpenKJ` (rcas 0.55) before `SaveImage` ships **active** — bypassing it

- drops the image link (a converter gap: bypass-passthrough doesn't cross a subgraph

- IMAGE output), and the contrast-adaptive sharpen suits Krea 2's crisp look anyway.

+ Both packs render crisp at 1920×1080 / 8 steps / cfg 1 / er_sde with the optional

+ loop nodes bypassed (default) — `-manual` on a prose snow-leopard prompt, `-json`

+ on a tea still-life whose teapot/cup/figs each land in their bbox region.

+ **Note:** the `ImageSharpenKJ` (rcas 0.55) before `SaveImage` is **active** —

+ bypassing it drops the image link (a converter gap: bypass-passthrough doesn't

+ cross a subgraph IMAGE output), and the contrast-adaptive sharpen suits Krea 2's

+ crisp look anyway.

## Gotchas

- `CLIPLoader: 'krea2' not in list` → ComfyUI too old; update to ≥ v0.26.0.

- `Torch not compiled with CUDA enabled` → reinstall torch for your CUDA tag

(`--index-url https://download.pytorch.org/whl/cu128`).