Merge 2bf3807408 into c4a14df9a3

Merge master into feature/qwen-image-edit-node
Resolved conflict in comfy_extras/nodes_qwen.py by keeping the alpha channel fix: - Retained the fix: images_vl.append(s.movedim(1, -1)[:, :, :, :3]) - This ensures only RGB channels are used for vision processing - Fixes RuntimeError when processing images with alpha channels
2026-02-05 19:12:41 +08:00 · 2026-01-21 01:36:55 +00:00 · 2025-09-25 14:24:13 +08:00 · 2025-09-25 13:33:35 +08:00 · 2025-09-15 14:00:33 +08:00 · 2025-09-15 13:53:09 +08:00
1 changed files with 1 additions and 1 deletions
--- a/comfy_extras/nodes_qwen.py
+++ b/comfy_extras/nodes_qwen.py
@ -87,7 +87,7 @@ class TextEncodeQwenImageEditPlus(io.ComfyNode):
                height = round(samples.shape[2] * scale_by)

                s = comfy.utils.common_upscale(samples, width, height, "area", "disabled")
-                images_vl.append(s.movedim(1, -1))
+                images_vl.append(s.movedim(1, -1)[:, :, :, :3])
                if vae is not None:
                    total = int(1024 * 1024)
                    scale_by = math.sqrt(total / (samples.shape[3] * samples.shape[2]))
Author	SHA1	Message	Date
AIGC_Zero	5002bff1ee	Merge `2bf3807408` into `c4a14df9a3`	2026-01-21 01:36:55 +00:00
AIGCZero	2bf3807408	Merge master into feature/qwen-image-edit-node Resolved conflict in comfy_extras/nodes_qwen.py by keeping the alpha channel fix: - Retained the fix: images_vl.append(s.movedim(1, -1)[:, :, :, :3]) - This ensures only RGB channels are used for vision processing - Fixes RuntimeError when processing images with alpha channels	2025-09-25 14:24:13 +08:00
AIGCZero	ffad732482	Fix alpha channel handling in Qwen image edit nodes - Fix TextEncodeQwenImageEditPlus to ensure only RGB channels are used - Prevents RuntimeError when input images have alpha channels - Ensures proper tensor shape for vision language models	2025-09-25 13:33:35 +08:00
AIGCZero	2debbcf50e	Remove Chinese comments from Qwen image edit node	2025-09-15 14:00:33 +08:00
AIGCZero	cc1c5da4f6	Add intelligent scaling algorithm selection for Qwen image edit node - Implement automatic algorithm selection: area for downscaling, lanczos for upscaling - Improve image quality by choosing optimal scaling method based on target size - Add Chinese comments for better code documentation - Ensure 8-pixel alignment for better compatibility with diffusion models	2025-09-15 13:53:09 +08:00
AIGCZero	c8bfec86b3	Add TextEncodeQwenImageEdit node with intelligent scaling - Implements Qwen image editing functionality with CLIP text encoding - Features intelligent scaling algorithm selection: - Uses 'area' method for downscaling to preserve details - Uses 'lanczos' method for upscaling for better quality - Supports optional VAE encoding for reference latents - Maintains aspect ratio with 'disabled' crop method - Scales images to target resolution (1024x1024 pixels) with 8-pixel alignment	2025-09-15 13:50:49 +08:00