Grok推出新圖像模型

2026年8月9日 00:00

重點摘要

Open Grok.com Imagine Image 2.0 is now generally available as the new Quality Mode on grok.com/imagine, and our iOS and Android apps. We built Image 2.

站內 AI 整理稿

Open Grok.com Imagine Image 2.0 is now generally available as the new Quality Mode on grok.com/imagine, and our iOS and Android apps.We built Image 2.0 around a simple goal: make images you can use in real work.2.0 follows instructions closely, down to the details.

It plans typography and layout the way a designer would, so dense, multi-part visuals hold together and small text comes out sharp.And it preserves what you put in across generations and edits.Precise editing Real work is iterative.The first generation is rarely the final asset, so Image 2.

0 adds tools for changing exactly what you mean and nothing else.The magic wand edits the region you point at and leaves the rest untouched.Segmentation selects precise areas of the image to change.Background removal exports any subject with a transparent background, ready to drop into other work.

Multi-ref editing accepts up to 5 input images in a single generation, removing the need for manual compositing.Smart resize One image, any size: pick a ratio and the model fills in the frame.1:21:29:162:33:41:14:33:216:92:1 Real-world performance Image 2.

0 was trained for fidelity across photography, design, and illustration, with editing treated as a first-class capability.It ranks second in the world in both text-to-image generation and image editing.Image Edit ArenaText-to-Image ArenaOverall Elo.

Source: Arena Image Edit and Text-to-Image leaderboards (as of Aug 7, 2026).xAI models are listed on Arena under SpaceXAI.Templates We’re also introducing new templates, which package common image workflows into ready-made starting points.

Each one covers a specific task — photo editing, product shots, headshots, icons, game assets, and more — with the workflow already configured, so you supply the inputs and get a finished result.

Photo ToolsPhoto EditProductProduct Color ChangeMarketingEditorial Product PosterPhoto ToolsReimaginePhoto ToolsPhoto CollageDesign ToolsMascot MakerPhoto ToolsBG Removal & ChangeMarketingE-Commerce PhotosMarketingUGC PhotosPhoto ToolsProfessional HeadshotDesign ToolsIcon MakerDesign ToolsCharacter SpriteGame AssetsProps & UI KitStreamingEmoji CreatorMarketingMerch Maker Build a world for video One world from a handful of prompts: a character, her locations, and the props she carries — generated separately, holding one style from image to image.

One worldBuilt image by imageA character, her locations, and the props she carries — generated separately, holding one look.CharacterLocationLocationPropPropCharacterLocation Try it today on our consumer products OpenGrok.comOpenGrok on iOSOpenGrok on Android API access is coming soon.

Related

相關文章

OpenClaw手機代理接入

OpenClaw手機代理接入。 作者把Open WebUI接上手機端。作者���在社區原文說明方案講清動機。它走WebSocket,工具卡實時流出。項目按MIT開放,兩文件可用。

4 小時前

谷歌技能倉庫升溫

谷歌技能倉庫升溫。 團隊公開了自動化工具集合代碼庫。開發者可以通過這個項目 ⚙️ 接入其核心接口。它能有效解決系統管理中的多種自動化痛點。該技能庫已經獲得了萬級以上的收藏關注。許多嚐鮮者反饋體驗 (✿◡‿◡) 非常流暢。

4 小時前

美國擬建 51 億美元數據中心遭擱置,當地激進民眾拿出斷頭臺以示抗議

首頁 > 智能時代>人工智能 美國擬建 51 億美元數據中心遭擱置,當地激進民眾拿出斷頭臺以示抗議 2026/8/8 13:09:18 來源:IT之家 作者:故淵 責編:故淵 評論: IT之家 8 月 8 日消息,科技媒體 Tom's Hardware 昨日(8 月 7 日)發佈博文,報道稱美國俄勒岡州塞勒姆市(Salem)關於擬建 51 億美元(IT之家注:現匯率約合 344.

15 小時前

秋招信息戰打不動?我們測了千問的新功能,讓Agent全程陪跑

阿里巴巴旗下大模型「千問」近日推出全新功能,針對秋季招聘季中求職者面臨的海量資訊與繁複流程,打造「Agent 全程陪跑」服務。根據官方測試,這項功能能協助使用者自動篩選職缺、整理企業資訊,甚至模擬面試問答,大幅減輕求職者自行蒐集與比對資訊的負擔。 在實際體驗中,用戶只需輸入自身學經歷與目標產業,Agent 便會自動爬取最新的秋招公告,並按職位匹配度、截止日期等條件排序,同時附上企業背景與筆面試經驗摘要。

1 天前