As competition for AI generative tools heats up and OpenAI recently released ChatGPT 5.1, xAI has rolled out a major update to Grok Image. The upgrade introduces image prompt editing, video preview and an HDR mode, evolving Grok Image from a static generator into an end‑to‑end creative engine. xAI founder Elon Musk posted on X: "Update your Grok App — edit image prompts easily now!" According to xAI's official announcement and the iOS App Store update, version v1.3.8 has been pushed to all users, with performance gains of over 20% in generation speed and consistency. TechCrunch reported the update addresses user feedback and improves prior limitations—such as the lack of iterative editing—signaling xAI's accelerated push in multimodal AI. While ChatGPT 5.1 emphasizes "humanized dialogue," Grok Image's upgrade focuses on creative productivity and is expected to further consolidate xAI's position in the image/video generation market.
Since the Grok Image series debuted with the Aurora model in August 2025, it has been a core part of the xAI ecosystem. Built on the Grok‑4 architecture, it supports text‑to‑image and image‑to‑image generation. The v1.3.8 update's highlights are UI and feature expansions that shift the tool from single‑shot generation to iterative creation. The official blog describes a new three‑column UI: content input on the left, voice/visual settings in the center, and generation preview on the right. Users can directly edit prompts via an "Edit" button—for example, iterating from "a cat in a forest" to "a cat chasing fireflies in a moonlit forest"—and the model regenerates in real time without retyping. An X user, @SidvexoAi, shared a UI preview showing a new "slide to view edit history" feature allowing horizontal browsing of variants; that post has over 500 likes. Compared with ChatGPT 5.1's "persona presets," Grok Image places greater emphasis on visual iteration, appealing to designers and content creators.
Performance improvements are the headline of this update. xAI's newsroom states Grok Image v1.3.8 achieves 150 tokens/s (A100 GPU baseline) with response times reduced to under 2 seconds (previously about 4 seconds), thanks to an optimized Aurora model and INT4 quantization (precision loss <1%). Hindustan Times reported the new image‑to‑video capability: users can upload a static image and generate a 5–10 second animated clip with effects like "object morphing" or "scene transitions." Tests indicate video consistency improved by 25% and character motion smoothness reached 92% (based on a LiveCodeBench variant). Compared with ChatGPT 5.1's voice I/O latency of 1.2 seconds, Grok Image's video generation emphasizes multimodal fusion—i10X called it a move from an image tool toward a full creative engine. The update also adds HDR/Normal mode switching; the HDR mode delivers up to 10‑bit color depth for professional photography output, and the preview feature lets users inspect video variants before saving.
Feature expansions further underline Grok Image's practicality. The new version integrates mute/play‑pause controls and an upgrade button enabling one‑click upscaling from 2K to 4K and sharing to X. Lance AI demonstrated editing prompts then sliding through history versions, with an interface described as as intuitive as Photoshop. Mirroring ChatGPT 5.1's persona options, Grok Image added "style presets" (for example, "Prada fashion shoot" or "Brandon Woelfel photography"); an example fashion portrait shared by user @AIRevSpot received over 1,000 likes. On metrics, character consistency rose to 95% (from 85%), and text‑rendering accuracy reached 98%, avoiding common issues like letter distortion seen in some models. CometAPI reported API call costs dropped to $0.039 per image (video +$0.05 per 5 seconds), offering competitive value relative to Midjourney. Elon Musk's post accumulated over 2.8 million views, urging users to "Update the App, edit prompts easily!"
Experts welcomed the upgrade. xAI engineers said the new Aurora model integrates sparse MoE routing to boost generation efficiency by 20% and supports up to 256K context windows for long image‑sequence narratives. DailyAIPrompt shared example prompts on X such as: "Edit [image]: change [element], keep the atmosphere." Compared with ChatGPT 5.1's 92% instruction‑follow accuracy, Grok Image's visual iteration consistency is reportedly higher, making it suitable for professional workflows like advertising design. i10X analysis projects a 20% user growth for xAI, challenging OpenAI's dominance in creative AI. However, The Register noted that while watermark detection has been strengthened (Synth‑ID anti‑screenshot), the open‑source community warns it may be possible to train methods to remove such marks within one to two years, requiring ongoing iteration.
The Grok Image upgrade could have broad effects on the AI ecosystem. Patrika reported that image‑to‑video capability shifts static creation toward dynamic forms, and estimated the generative video market could reach $100 billion by 2026. In contrast to ChatGPT 5.1's focus on "warm dialogue," Grok Image emphasizes a "visual partner" approach; xAI's blog indicates plans to integrate Grok Code Remote for cloud‑based code‑to‑image generation. Some users, like Aegis999 on X, criticized earlier overly strict moderation and say the new version feels more permissive. Shannon AI discussed Grok Image detection tools in a post that received significant attention. Overall, the update reinforces xAI's "open + practical" positioning—Elon Musk's post views signal an influx of users.
The near‑simultaneous releases of ChatGPT 5.1 and Grok Image's upgrade highlight the transition from single‑modality tools to multimodal ecosystems. OpenAI's 5.1 focuses on more natural dialogue (1.2s latency, ~92% accuracy), while xAI's Grok Image increases visual production capacity (150 tokens/s, ~95% consistency). Decrypt predicts the multimodal AI market could exceed $200 billion by 2026; both products will set benchmarks. Experts such as Percy Liang described Grok Image's editing fluidity as "Photoshop AI," while ChatGPT 5.1's presets were likened to "Siri 2.0." Amid a global AI investment surge, these dual advances may accelerate innovation but also raise ethical concerns around misuse. Looking ahead, xAI may launch Grok Video and OpenAI may expand visual modes, as AI creative tools reshape entertainment and education.