Advanced reference

Krea2 Edit Technical Guide

This page collects the model, tuning, and local-workflow details behind the browser editor. It is written for users who want pinned weights, repeatable settings, lower-VRAM variants, or a local ComfyUI graph.

Model weight and version guidance

Prefer v1.2 when it is available. It is the recommended build on the model card and adds reference sheets, swaps, inpainting, outpainting, virtual try-on, and FIT geometry. Confirm the version pinned by the hosted Space before comparing results across sessions.

Recommended file

krea2_identity_edit_v1_2.safetensors

更改 Krea2 Edit 結果的設置

Treat these values as starting points from the public model card and node changelog, not universal guarantees. Change one control at a time and compare against the source image.

職位類型檢查站步驟CFG
新增/重新著色/重新設計/重新舞台Krea 2 Turbo8–121.0
刪除/大量刪除Krea 2 Raw~203.0
  • ref_boost ≈ 4 — v1.2 節點上的強烈相似起點;高於 ~10 開始破壞移除;低於 1 會削弱創造力。
  • grounding_px — 較低的版本更一致地遵循編輯;較高的保護身分。重複的「分割圖片」影格通常表示該值遠高於訓練範圍(v1.1 文件引用 384-768 作為舒適帶)。
  • LoRA strength 1.0 — 已發布預設值。當 CFG 高於 1 時,將負片與空提示加上相同的影像接地。
  • v1.2 的低 VRAM SVD 刪減( r128 ~ 0.91 GB, r64 ~ 0.46 GB) 每張卡保留 >99% 的重量能量 — 當您將瀏覽器 Krea2 Edit 留給筆記型電腦 GPU 時非常有用。

瀏覽器 Krea2 Edit 與本機 ComfyUI

The hosted editor is best for quick prompt tests. A local graph is better when you need exact versions, repeatable seeds, batches, private infrastructure, or additional LoRAs.

託管 Space(本頁)

  • 無檢查點下載;在幾分鐘內嘗試 Krea2 Edit
  • 適合 Krea2 Edit 上的快速研發和客戶端演示
  • 取決於 Space 正常運作時間和共享 GPU 佇列
  • 全螢幕開啟: hf.space

本地 ComfyUI-Krea2Edit

  • 腳位 v1.2 權重與 FIT / ref_boost 節點版本
  • 在頂部堆疊字元、內文或樣式 LoRA
  • 需要原生 Krea 2 + Qwen3-VL 編碼器支持
  • ComfyUI-Krea2Edit on GitHub

Krea2 Edit 的限制、授權和負責任的使用

Krea2 Edit 上的相似性在上游被描述為紋理忠實且比例保守:皮膚和燈光通常轉移得很好,而非常不尋常的面部幾何形狀可能看起來像近親。服裝交換是偶然的;刪除需要 Raw 配方,但有時仍然會失敗。極端的頭髮或身體類型可能需要在頂部放置主題 LoRA。

權重是下式下的導數模型 Krea 2 Community License。在公佈的收入門檻下允許商業使用(模型卡摘要:低於 100 萬美元/年);除此之外,請聯絡 Krea 以了解企業條款。部署應適度內容; AI 揭露可能適用。 LoRA 經過 SFW 訓練 - 不會創造未經同意或有害的真人肖像。閱讀我們的 隱私權政策 服務條款.

Technical sources

Check the upstream model card and repositories for current files, workflows, changelogs, licenses, and compatibility notes.

FAQ

Technical FAQ

Version and local-workflow answers for advanced users.

Which model weight should I use?

Prefer v1.2 (krea2_identity_edit_v1_2.safetensors) when available, and confirm the exact version pinned by the hosted Space before comparing sessions.

What do I need for a local workflow?

Use ComfyUI with native Krea 2 and Qwen3-VL support, install the ComfyUI-Krea2Edit node pack, load the LoRA, and start from a workflow shipped by the project.

What settings should I start with?

For common additions, recolors, restyles, and restages, start with the Turbo recipe at 8–12 steps and CFG 1.0. For large removals, try the Raw recipe near 20 steps and CFG 3.0.