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。

Check out Lumina-Image 2.0 🖼️ an efficient, unified, and transparent image generative model built with Gemma 2's text encoder and FLUX’s VAE 👇

Google AI Developers

114,024 subscribers

33,467 次观看 • 1 年前 •via X (Twitter)

科学技术教育艺术

Anya Rossi• Live Now

Private livecam show

8 条评论

Google AI Developers 的头像

Google AI Developers1 年前

The model is multilingual, can handle system prompts, and packs a punch for its relatively small parameter size.

Google AI Developers 的头像

Google AI Developers1 年前

It’s a great example of how vision models, and more, can be built on top of Gemma 2’s text encoder.

Google AI Developers 的头像

Google AI Developers1 年前

Access the model here and let us know what you’re building with Gemma in the comments

Matt Figdore 的头像

Matt Figdore2 年前

This is the biggest productivity cheat code right now. Kiss reading documents goodbye. You can get an instant summary of any document with this tool.

TestingCatalog News 🗞 的头像

TestingCatalog News 🗞1 年前

Should be Imagemma 2.0 😁 Let’s test 👀

L8NTLABS 的头像

L8NTLABS1 年前

Lumina-Image 2.0 sounds like a game changer, especially with the integration of Gemma 2's text encoder and FLUX's VAE. I'd love to see some examples of what it can produce. Can you describe one of the images generated by Lumina-Image 2.0?

Chris Mur l 的头像

Chris Mur l1 年前

𝗔𝘂𝘁𝗼 𝗥𝗲𝗽𝗹𝘆 Can't talk now. 𝘞𝘩𝑎𝑡𝑎𝑢𝑡𝑜

⟁ndrew V 的头像

⟁ndrew V1 年前

wow great work!!!

相关视频

Designing an Encoder for Fast Personalization of Text-to-Image Models TL;DR: use an encoder to personalize a text-to-image model to new concepts with a single image and 5-15 tuning steps abs: project page:

Designing an Encoder for Fast Personalization of Text-to-Image Models TL;DR: use an encoder to personalize a text-to-image model to new concepts with a single image and 5-15 tuning steps abs: project page:

AK

165,158 次观看 • 3 年前

New Kling, Runway, Luma competitor? Open text and image video generation model Pyramidal Flow Matching for Efficient Video Generative Modeling

New Kling, Runway, Luma competitor? Open text and image video generation model Pyramidal Flow Matching for Efficient Video Generative Modeling

AK

90,965 次观看 • 1 年前

Gemma 4 12B dropped today. Apache 2.0, multimodal: text, image, audio, and video. 256K context, built-in thinking, native tool calling. Running on Red Hat OpenShift AI with vLLM on Day 0:

Gemma 4 12B dropped today. Apache 2.0, multimodal: text, image, audio, and video. 256K context, built-in thinking, native tool calling. Running on Red Hat OpenShift AI with vLLM on Day 0:

Red Hat AI

15,950 次观看 • 1 个月前

Qwen‑Image‑2.0 continues to push what’s possible in next‑gen image generation. From pro typography rendering and stronger semantic adherence to improved text rendering and a lighter model, it keeps delivering creative power with ease. Check out the demos to discover more!

Qwen‑Image‑2.0 continues to push what’s possible in next‑gen image generation. From pro typography rendering and stronger semantic adherence to improved text rendering and a lighter model, it keeps delivering creative power with ease. Check out the demos to discover more!

Alibaba Cloud

928,486 次观看 • 3 个月前

GPT Image 2.0 lands on Higgsfield. 🧩 OpenAI delivers perfect text rendering and ChatGPT-level reasoning in its SOTA image model. Open Higgsfield and generate.

GPT Image 2.0 lands on Higgsfield. 🧩 OpenAI delivers perfect text rendering and ChatGPT-level reasoning in its SOTA image model. Open Higgsfield and generate.

Higgsfield AI 🧩

1,535,120 次观看 • 2 个月前

gpt-image-2 is here, available today in the API and Codex. The most capable image generation model yet, built for production-grade workflows with stronger text rendering, layout, editing, resolution, and multilingual rendering.

gpt-image-2 is here, available today in the API and Codex. The most capable image generation model yet, built for production-grade workflows with stronger text rendering, layout, editing, resolution, and multilingual rendering.

OpenAI Developers

181,551 次观看 • 2 个月前

InfinityStar by Bytedance: A unified 8B spacetime autoregressive model for high-res image & video gen; - 5s 720p video ~10x faster than DiT; - scores 83.74 on VBench, topping other AR models and HunyuanVideo; - Flan-T5-XL as text encoder. - 480/720p, ~35Gb model

InfinityStar by Bytedance: A unified 8B spacetime autoregressive model for high-res image & video gen; - 5s 720p video ~10x faster than DiT; - scores 83.74 on VBench, topping other AR models and HunyuanVideo; - Flan-T5-XL as text encoder. - 480/720p, ~35Gb model

Wildminder

23,255 次观看 • 8 个月前

Google is offering a Generative AI Learning Path with 10 courses for FREE! - Intro to Generative AI - Intro to LLMs - Intro to Image Generation - Encoder-Decoder Architecture - Transformer Models and more A Thread 🧵👇

Google is offering a Generative AI Learning Path with 10 courses for FREE! - Intro to Generative AI - Intro to LLMs - Intro to Image Generation - Encoder-Decoder Architecture - Transformer Models and more A Thread 🧵👇

Afiz ⚡️

249,552 次观看 • 3 年前

Struggling with slow inference of diffusion and flow models? Check out the video below—I’ve been using our new FastGen library to achieve 7-28x acceleration for text-2-image and {text,image,video}-2-video generation without sacrificing visual fidelity!

Struggling with slow inference of diffusion and flow models? Check out the video below—I’ve been using our new FastGen library to achieve 7-28x acceleration for text-2-image and {text,image,video}-2-video generation without sacrificing visual fidelity!

Julius Berner

13,643 次观看 • 4 个月前

Image. Then motion. GPT Image 2 + Seedance 2.0 are now both live on Pollo AI. GPT Image 2 → structured layouts, sharp text, consistent characters — ready for ads, products, and social content. Seedance 2.0 → cinematic, stable, smooth video — straight from your image. From image to motion, the workflow is fast, seamless, and powerful.

Image. Then motion. GPT Image 2 + Seedance 2.0 are now both live on Pollo AI. GPT Image 2 → structured layouts, sharp text, consistent characters — ready for ads, products, and social content. Seedance 2.0 → cinematic, stable, smooth video — straight from your image. From image to motion, the workflow is fast, seamless, and powerful.

Pollo AI

431,569 次观看 • 2 个月前

Phidias A Generative Model for Creating 3D Content from Text, Image, and 3D Conditions with Reference-Augmented Diffusion discuss: In 3D modeling, designers often use an existing 3D model as a reference to create new ones. This practice has inspired the development of Phidias, a novel generative model that uses diffusion for reference-augmented 3D generation. Given an image, our method leverages a retrieved or user-provided 3D reference model to guide the generation process, thereby enhancing the generation quality, generalization ability, and controllability. Our model integrates three key components: 1) meta-ControlNet that dynamically modulates the conditioning strength, 2) dynamic reference routing that mitigates misalignment between the input image and 3D reference, and 3) self-reference augmentations that enable self-supervised training with a progressive curriculum. Collectively, these designs result in a clear improvement over existing methods. Phidias establishes a unified framework for 3D generation using text, image, and 3D conditions with versatile applications.

Phidias A Generative Model for Creating 3D Content from Text, Image, and 3D Conditions with Reference-Augmented Diffusion discuss: In 3D modeling, designers often use an existing 3D model as a reference to create new ones. This practice has inspired the development of Phidias, a novel generative model that uses diffusion for reference-augmented 3D generation. Given an image, our method leverages a retrieved or user-provided 3D reference model to guide the generation process, thereby enhancing the generation quality, generalization ability, and controllability. Our model integrates three key components: 1) meta-ControlNet that dynamically modulates the conditioning strength, 2) dynamic reference routing that mitigates misalignment between the input image and 3D reference, and 3) self-reference augmentations that enable self-supervised training with a progressive curriculum. Collectively, these designs result in a clear improvement over existing methods. Phidias establishes a unified framework for 3D generation using text, image, and 3D conditions with versatile applications.

AK

25,120 次观看 • 1 年前

🧵 One Image, Six Unique Videos: All with Text Prompts 🎥 One image. Six distinct stories. All by tweaking prompts. I've been loving the new Hailuo 02 video model, and its ability to follow instructions is mind-blowing. Check out these 6 examples ... 👇

🧵 One Image, Six Unique Videos: All with Text Prompts 🎥 One image. Six distinct stories. All by tweaking prompts. I've been loving the new Hailuo 02 video model, and its ability to follow instructions is mind-blowing. Check out these 6 examples ... 👇

Marco

40,864 次观看 • 1 年前

Wan2.6: Commercial-Grade Image Output Wan2.6-Image is now available. 🖼️ Interleaved Text-and-Image Output：Generate interleaved text-and-image content with logical reasoning capabilities — enabling layered, narrative-driven visual storytelling. 🖼️ Multi-Image Conditioned Generation：Support flexible referencing, combining, and replacement of multiple images, integrating varied visual inspirations to generate novel and compelling results 🖼️ Commercial-Grade ID Preservation：High consistency in characters, styles, and elements for commercial scenarios. 🖼️ Extract creative elements — such as color, style, and composition — from reference images to enable aesthetically driven image generation. 🖼️ Precise Control of Camera Angles and Lighting：Support specifying camera perspective, spatial depth (foreground/background), and lighting details — enabling precise control over spatial composition and atmospheric mood. Try Wan2.6 today.

Wan2.6: Commercial-Grade Image Output Wan2.6-Image is now available. 🖼️ Interleaved Text-and-Image Output：Generate interleaved text-and-image content with logical reasoning capabilities — enabling layered, narrative-driven visual storytelling. 🖼️ Multi-Image Conditioned Generation：Support flexible referencing, combining, and replacement of multiple images, integrating varied visual inspirations to generate novel and compelling results 🖼️ Commercial-Grade ID Preservation：High consistency in characters, styles, and elements for commercial scenarios. 🖼️ Extract creative elements — such as color, style, and composition — from reference images to enable aesthetically driven image generation. 🖼️ Precise Control of Camera Angles and Lighting：Support specifying camera perspective, spatial depth (foreground/background), and lighting details — enabling precise control over spatial composition and atmospheric mood. Try Wan2.6 today.

Wan

33,264 次观看 • 6 个月前

🎇 Video Model Comparison: Image2Video Video Model Comparison: Image2Video Same input image + text prompt on each model: • Pika 2.0 • PixVerse 3.5 • Runway Gen-3 • Kling AI 1.6 • Luma Dream Machine • Hailuo MiniMax I used the same Midjourney image and text prompt in each model and chose my favorite results.

🎇 Video Model Comparison: Image2Video Video Model Comparison: Image2Video Same input image + text prompt on each model: • Pika 2.0 • PixVerse 3.5 • Runway Gen-3 • Kling AI 1.6 • Luma Dream Machine • Hailuo MiniMax I used the same Midjourney image and text prompt in each model and chose my favorite results.

Heather Cooper

36,110 次观看 • 1 年前

Gemma 4 31B is now available in Public Preview on Cerebras. Our first multimodal model runs at over 1,800 tokens/s for ultra-fast image and text workflows. Give it a try:

Gemma 4 31B is now available in Public Preview on Cerebras. Our first multimodal model runs at over 1,800 tokens/s for ultra-fast image and text workflows. Give it a try:

Cerebras

162,940 次观看 • 6 天前

The new 2026 AI workflow 🤖⚡ GPT Image 2.0 → assets → Codex + GPT-5.5 → playable browser game 🎮 I turned my AI image into a web game. 🖼️ GPT Image 2 created the visual. 🎨 Codex generated the sprites and animations. 🕹️ GPT-5.5 built the prototype.

The new 2026 AI workflow 🤖⚡ GPT Image 2.0 → assets → Codex + GPT-5.5 → playable browser game 🎮 I turned my AI image into a web game. 🖼️ GPT Image 2 created the visual. 🎨 Codex generated the sprites and animations. 🕹️ GPT-5.5 built the prototype.

Givros

941,670 次观看 • 2 个月前

Apple just released and open-sourced FastVLM! FastVLM is a lightning-fast vision-language model that combines rapid image and text understanding with efficient on-device performance. 100% Open Source

Apple just released and open-sourced FastVLM! FastVLM is a lightning-fast vision-language model that combines rapid image and text understanding with efficient on-device performance. 100% Open Source

Sumanth

43,704 次观看 • 10 个月前

Introducing ImagineArt 2.0 - The world's most realistic image generation model! 🧠 Reasoning-based prompt adherence 🎬 Cinematic photography ⚡ Concept to final in seconds 🔤 Text rendering that actually works This is what happens when an image model stops generating and starts reasoning.

Introducing ImagineArt 2.0 - The world's most realistic image generation model! 🧠 Reasoning-based prompt adherence 🎬 Cinematic photography ⚡ Concept to final in seconds 🔤 Text rendering that actually works This is what happens when an image model stops generating and starts reasoning.

ImagineArt

714,542 次观看 • 2 个月前

𝐒𝐔𝐏𝐈𝐑: An advanced image restoration method that combines generative prior and model scaling,using a 20-million image dataset, and can manipulate restorations with textual prompts. Kudos to Fanghua Yu Jinjin Gu Xintao Wang et al🚀 SUPIR supports Gradio demo in repo [links👇]

𝐒𝐔𝐏𝐈𝐑: An advanced image restoration method that combines generative prior and model scaling,using a 20-million image dataset, and can manipulate restorations with textual prompts. Kudos to Fanghua Yu Jinjin Gu Xintao Wang et al🚀 SUPIR supports Gradio demo in repo [links👇]

Gradio

56,351 次观看 • 2 年前

Text to image with midjourney and image to video with gen2 by @commonstyle

Text to image with midjourney and image to video with gen2 by @commonstyle

AK

668,122 次观看 • 2 年前