正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

Ideogram 4.5 is now in ComfyUI. It's an edit model built to fix drift. 4.5 keeps the rest of your image intact, edit after edit. → Multi-turn edits that stay clean → Edit 24MP+ images without downscaling, with crop edges preserved so the edit stitches back in seamlessly →... show more

ComfyUI

60,147 subscribers

20,889 次观看 • 20 小时前 •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

23 条评论

ComfyUI 的头像

ComfyUI20 小时前

Ideogram 4.5 Precise Edit

ComfyUI 的头像

ComfyUI20 小时前

Ideogram 4.5: Image Edit

ComfyUI 的头像

ComfyUI20 小时前

Ideogram 4.5: Text to Image

2nd Harvest 的头像

2nd Harvest20 小时前

*Insert generic AI response repeating what the original post already says but from an account with a cute white chick profile pic.

Brie Wensleydale🧀🐭 的头像

Brie Wensleydale🧀🐭20 小时前

No models yet it seems ~

Sophia AI & Tool Expert 的头像

Sophia AI & Tool Expert15 小时前

24MP with no downscaling crop edges that stitch back seamlessly is insane for Comfy workflows finally multi turn edits that dont punish you

روم به دیوار 🗿 的头像

روم به دیوار 🗿12 小时前

How many refrence images it can accept?

Claire Bennett 的头像

Claire Bennett20 小时前

Ideogram 4.5 looks like a strong upgrade for precise image editing. Keeping the original image intact across multiple edits while handling high-resolution assets is a huge advantage for professional workflows.

BURLY STUDIO AI 的头像

BURLY STUDIO AI8 小时前

Ideogram 4.5 is amazing

Aria Tech 的头像

Aria Tech16 小时前

Fantastic integration perfect for high res workflows

玻璃渣的头像

玻璃渣7 小时前

fixing drift across edits is the real unlock here

Chole Syntax Expert AI 的头像

Chole Syntax Expert AI15 小时前

Ideogram 4.5 fixing drift is a game changer multi turn edits staying clean at 24MP is exactly what ComfyUI needed

NOOR TECH 的头像

NOOR TECH15 小时前

Ideogram 4.5 in ComfyUI fixes drift, keeps images intact across multi-turn edits, preserves 24MP+ crops seamlessly clean always

ziziran 的头像

ziziran19 小时前

局部编辑有个挺有意思的边界：商品换了颜色，镜子里的倒影也得一起变。商品和倒影相隔很远时，这套裁切回贴工作流怎么联动更新两处，同时保住中间的背景？

Elara AI 的头像

Elara AI16 小时前

Multi turn edits staying clean is exactly what we needed

A.W.E.S.O.M.-O 4000 的头像

A.W.E.S.O.M.-O 40008 小时前

ngl ideogram 4.5 in comfyui is lowkey the combo i was waiting for

Somya Singh 的头像

Somya Singh13 小时前

Stitching crops back seamlessly might fail if an edit changes local lighting, creating a visible seam against the unedited background. How does the model handle global illumination consistency across those preserved crop edges?

Vera Ai | Tools & Updates 的头像

Vera Ai | Tools & Updates12 小时前

Keeping image consistency across edits is a huge workflow upgrade

Aina Ai | Tools & Updates 的头像

Aina Ai | Tools & Updates15 小时前

No more drift flawless multi edits at 24MP

Alice The Ai Expert 的头像

Alice The Ai Expert16 小时前

Perfect ComfyUI + Ideogram 4.5 is the combo we needed for clean, drift free edits at full resolution!

Vantix AI Agency 的头像

Vantix AI Agency14 小时前

Edit without drift 24MP without downscaling is massive. Multi turn staying clean plus seamless stitch back Ideogram 4.5 in ComfyUI is exactly what pros needed

Wassollichhier 的头像

Wassollichhier8 小时前

@ideogram_ai Is image to image possible? Virtual try on?

Amelia | AI & Tech 的头像

Amelia | AI & Tech11 小时前

Finally an edit model that actually preserves the original

相关视频

Seedream 5.0 Pro is now in ComfyUI via Partner Nodes. ByteDance's latest image model brings: → Character & product consistency across edits → Precision region editing → Infographics, flowcharts & structured layouts → In-image text across 14 languages One model. Text-to-image, editing, and multi-image inputs. Click the link below for the workflow 👇

Seedream 5.0 Pro is now in ComfyUI via Partner Nodes. ByteDance's latest image model brings: → Character & product consistency across edits → Precision region editing → Infographics, flowcharts & structured layouts → In-image text across 14 languages One model. Text-to-image, editing, and multi-image inputs. Click the link below for the workflow 👇

ComfyUI

14,756 次观看 • 2 个月前

🔥 This Flint Image Edit. Finally, the edit model you wanted, but cheaper. Still editing one version at a time? Flint Image Edit takes up to ten reference images and one prompt, and returns four edited versions in a single call.

🔥 This Flint Image Edit. Finally, the edit model you wanted, but cheaper. Still editing one version at a time? Flint Image Edit takes up to ten reference images and one prompt, and returns four edited versions in a single call.

each::labs

306,509 次观看 • 2 个月前

Precise Edit is now available on Grok Imagine. It allows you to easily edit a specific part of an image without changing the rest. Grok automatically splits your image into multiple segments that you can select and edit individually. A super useful upgrade. Try it now!

Precise Edit is now available on Grok Imagine. It allows you to easily edit a specific part of an image without changing the rest. Grok automatically splits your image into multiple segments that you can select and edit individually. A super useful upgrade. Try it now!

DogeDesigner

68,901 次观看 • 1 个月前

The new Qwen-Image-Edit-2509 is now live! Edit multiple images, enjoy greatly enhanced consistency for faces and text, and leverage native ControlNet support. Explore the update in the "Image Editing" feature of Qwen Chat.

The new Qwen-Image-Edit-2509 is now live! Edit multiple images, enjoy greatly enhanced consistency for faces and text, and leverage native ControlNet support. Explore the update in the "Image Editing" feature of Qwen Chat.

Tongyi Lab

235,327 次观看 • 1 年前

🔥Touch Edit: The Next Level of Image edit Already used by 100,000+ creators, this feature elevates Lovart to a new degree of Design. Mark & Edit: Point to specific details (clothes, text, objects) in the image, mark the area, and edit with perfect accuracy. Select & Remix: Fuse elements from multiple images into one seamless masterpiece, all in a single prompt

🔥Touch Edit: The Next Level of Image edit Already used by 100,000+ creators, this feature elevates Lovart to a new degree of Design. Mark & Edit: Point to specific details (clothes, text, objects) in the image, mark the area, and edit with perfect accuracy. Select & Remix: Fuse elements from multiple images into one seamless masterpiece, all in a single prompt

LovartAI

4,174,205 次观看 • 10 个月前

FLUX 3 Video is live in ComfyUI via Partner Nodes Black Forest Labs FLUX 3 Model highlights: → 20s in a single generation → native audio, not dubbed on after → text-to-video, image-to-video, keyframes, continuation → multi-scene and multi-angle in one prompt → multilingual dialogue with lipsync that holds → typography that sits in the scene → chain clips agentically for longer stories One model across video, audio, image, and action. Click the link below to download the workflows and learn more with blog

FLUX 3 Video is live in ComfyUI via Partner Nodes Black Forest Labs FLUX 3 Model highlights: → 20s in a single generation → native audio, not dubbed on after → text-to-video, image-to-video, keyframes, continuation → multi-scene and multi-angle in one prompt → multilingual dialogue with lipsync that holds → typography that sits in the scene → chain clips agentically for longer stories One model across video, audio, image, and action. Click the link below to download the workflows and learn more with blog

ComfyUI

25,715 次观看 • 1 个月前

3D editing is hard: you need to ground an image + instruction and generate a faithful 3D shape in one forward pass -- no test-time optimization. So, we steer pretrained image-to-3D representations to do text-guided 3D edits; no massive 3D edit-pair dataset needed. Key trap: the “no-edit” solution is a nasty local minimum. We fix it with preference optimization, pushing the model to actually edit. Steer3D is the second work that adapts alignment ideas from LLMs to the 3D modality. SAM 3D also used DPO to improve its 3D generations.

3D editing is hard: you need to ground an image + instruction and generate a faithful 3D shape in one forward pass -- no test-time optimization. So, we steer pretrained image-to-3D representations to do text-guided 3D edits; no massive 3D edit-pair dataset needed. Key trap: the “no-edit” solution is a nasty local minimum. We fix it with preference optimization, pushing the model to actually edit. Steer3D is the second work that adapts alignment ideas from LLMs to the 3D modality. SAM 3D also used DPO to improve its 3D generations.

Georgia Gkioxari

116,203 次观看 • 9 个月前

You can now generate, edit, and convert a reference image into 3D inside one tool. Tripo's Image Generation runs Nano Banana and GPT to build the reference, then turns it into a model. Text or photo in, 3D out. One pipeline, zero handoffs.

You can now generate, edit, and convert a reference image into 3D inside one tool. Tripo's Image Generation runs Nano Banana and GPT to build the reference, then turns it into a model. Text or photo in, 3D out. One pipeline, zero handoffs.

Alvaro Cintas

56,246 次观看 • 3 个月前

🚨🍌Big Reveal: who was "Nano Banana?" The anonymous model, “nano-banana,” that caught the world's attention with its ability to follow complex instructions, preserve character identity, and maintain contextual details was: Gemini-2.5-Flash-Image-Preview by Google DeepMind 🍌✨ - Now ranked #1 on the Image Edit Arena - Also ranked #1 for Text-to-Image In two weeks, “nano-banana” has driven over 5 million votes to the Image Edit Arena. With 2.5M+ votes for this model, it is the highest number of votes any model has received, with the largest Elo score lead (171) in Arena history. Congrats to the Google DeepMind team on this incredible milestone in image edit and generation. 👏

🚨🍌Big Reveal: who was "Nano Banana?" The anonymous model, “nano-banana,” that caught the world's attention with its ability to follow complex instructions, preserve character identity, and maintain contextual details was: Gemini-2.5-Flash-Image-Preview by Google DeepMind 🍌✨ - Now ranked #1 on the Image Edit Arena - Also ranked #1 for Text-to-Image In two weeks, “nano-banana” has driven over 5 million votes to the Image Edit Arena. With 2.5M+ votes for this model, it is the highest number of votes any model has received, with the largest Elo score lead (171) in Arena history. Congrats to the Google DeepMind team on this incredible milestone in image edit and generation. 👏

Arena.ai

106,535 次观看 • 1 年前

Hy Image3.5 preview is now available in ComfyUI. Professional-grade image generation, +30% win rate in human eval vs Hy Image3.0 → Text to image and Image to image in one model, up to 2K → Multilingual text, symbols, and small print that render correctly → Cinematic, comic, commercial photography, and illustration styles → Identity and product features that hold through scene, outfit, and style changes

Hy Image3.5 preview is now available in ComfyUI. Professional-grade image generation, +30% win rate in human eval vs Hy Image3.0 → Text to image and Image to image in one model, up to 2K → Multilingual text, symbols, and small print that render correctly → Cinematic, comic, commercial photography, and illustration styles → Identity and product features that hold through scene, outfit, and style changes

ComfyUI

26,315 次观看 • 9 天前

✨ NEW: Feature update: 🖼️ Image edit models now support multi-turn editing! Instead of trying to fit every edit into one mega-prompt, you can now refine your image step by step. Like a natural back-and-forth conversation. Do it in Battle mode, Side by Side or Direct. Just be sure to have the Image modality selected.

✨ NEW: Feature update: 🖼️ Image edit models now support multi-turn editing! Instead of trying to fit every edit into one mega-prompt, you can now refine your image step by step. Like a natural back-and-forth conversation. Do it in Battle mode, Side by Side or Direct. Just be sure to have the Image modality selected.

arena.ai

20,652 次观看 • 1 年前

Qwen-Image-2.1 is now supported in ComfyUI! Open weights. One 7B checkpoint that generates and edits. → Image generation at native 2K → Instruction editing from up to 10 reference images in a single pass → RGBA output, alpha included

Qwen-Image-2.1 is now supported in ComfyUI! Open weights. One 7B checkpoint that generates and edits. → Image generation at native 2K → Instruction editing from up to 10 reference images in a single pass → RGBA output, alpha included

ComfyUI

191,071 次观看 • 11 天前

One of our most requested features is here! ✍️✨ Introducing Direct Edits. Vibe Design is incredible for exploring ideas at lightspeed, but sometimes you just need to fix a typo or swap an image. Now, you can add that final layer of polish right inside Stitch. You can manually edit text and images, or highlight a specific part of the screen and ask the agent to surgically update it. 3 ways to jump in: 1️⃣ Select a screen and hit the Edit dropdown. 2️⃣ Grab the Edit Tool from the global toolbar. 3️⃣ Select Tool: Double-click a screen to center it, then double-click again to enter Direct Edit mode. Go perfect those vibes. ✨

One of our most requested features is here! ✍️✨ Introducing Direct Edits. Vibe Design is incredible for exploring ideas at lightspeed, but sometimes you just need to fix a typo or swap an image. Now, you can add that final layer of polish right inside Stitch. You can manually edit text and images, or highlight a specific part of the screen and ask the agent to surgically update it. 3 ways to jump in: 1️⃣ Select a screen and hit the Edit dropdown. 2️⃣ Grab the Edit Tool from the global toolbar. 3️⃣ Select Tool: Double-click a screen to center it, then double-click again to enter Direct Edit mode. Go perfect those vibes. ✨

Stitch by Google

187,600 次观看 • 7 个月前

Image Edit is heating up in the Arena - 3 new models have been added! ✨ Flux-Kontext-Max by BFL ✨ Bagel by ByteDance ✨ Step1X-edit by StepFun This brings the Image Edit Arena to a total of 7, with more coming! Upload an image and test them out, let's see what you think of these new models🖼️

Image Edit is heating up in the Arena - 3 new models have been added! ✨ Flux-Kontext-Max by BFL ✨ Bagel by ByteDance ✨ Step1X-edit by StepFun This brings the Image Edit Arena to a total of 7, with more coming! Upload an image and test them out, let's see what you think of these new models🖼️

Arena.ai

18,500 次观看 • 1 年前

Qwen Image 3.0 Pro is live in ComfyUI via Partner Nodes. → 4,500-token prompts (4.5x the last gen) → Legible text down to ~10px → 12 languages · 20+ fonts · 100+ styles → Formulas, diagrams, nested UI mockups → Multi-image editing Qwen is turning image generation into a tool for posters, slides, infographics, and product design. Click the link below to try it today 👇

Qwen Image 3.0 Pro is live in ComfyUI via Partner Nodes. → 4,500-token prompts (4.5x the last gen) → Legible text down to ~10px → 12 languages · 20+ fonts · 100+ styles → Formulas, diagrams, nested UI mockups → Multi-image editing Qwen is turning image generation into a tool for posters, slides, infographics, and product design. Click the link below to try it today 👇

ComfyUI

24,906 次观看 • 1 个月前

🎯Touch Edit just landed. Pinpoint Precision: Point to specific details (clothes, text, objects) in the image to edit with perfect accuracy. Select & Remix: Fuse elements from multiple images into one seamless masterpiece, all in one prompt. #Lovart #FeatureUpdate

🎯Touch Edit just landed. Pinpoint Precision: Point to specific details (clothes, text, objects) in the image to edit with perfect accuracy. Select & Remix: Fuse elements from multiple images into one seamless masterpiece, all in one prompt. #Lovart #FeatureUpdate

LovartAI

138,823 次观看 • 10 个月前

SeedEdit: ByteDance’s Open-Source Photoshop Killer It is an AI-powered image editor that’s completely free (link in the comments). With this tool, you can edit images effortlessly using simple text prompts. Here’s what you can do: - Change facial expressions and poses - Edit text within images - Seamlessly swap backgrounds - Adjust lighting and restyle scenes 9 examples:

SeedEdit: ByteDance’s Open-Source Photoshop Killer It is an AI-powered image editor that’s completely free (link in the comments). With this tool, you can edit images effortlessly using simple text prompts. Here’s what you can do: - Change facial expressions and poses - Edit text within images - Seamlessly swap backgrounds - Adjust lighting and restyle scenes 9 examples:

el.cine

167,814 次观看 • 1 年前

Aleph 2.0 is here. Now you can edit a single frame in your video, preview the change and then Aleph 2.0 carries that edit across the rest of your video. Try it now in the new Edit Studio on web at the link below.

Aleph 2.0 is here. Now you can edit a single frame in your video, preview the change and then Aleph 2.0 carries that edit across the rest of your video. Try it now in the new Edit Studio on web at the link below.

Runway

4,957,150 次观看 • 4 个月前