分类筛选
找到 731 个 Demo
Rate my personal website I built using Google AI S...
Rate my personal website I built using Google AI Studio 👀 I’m about to giveaway a personal website to 1 user who: • follows me • likes this post (ends in 48 hours ⏳) Today I built a website for myself just for fun but I actually loved the output. Prompting is the magic behind any good website. If you want to build the same kind of premium website for yourself, drop a reply with “prompt” and I’ll share the exact prompt I used.
Someone just built a real time- intelligence syste...
Someone just built a real time- intelligence system with Gemini 3.1 and Claude 4.6. https://t.co/8Pa6VneEJ4
The gap between "enterprise software" and "weekend...
The gap between "enterprise software" and "weekend project" just collapsed 🤯 Someone built a Google Earth + Palantir hybrid with real-time plane tracking + traffic cams... in ONE DAY. Using Gemini 3.1 + Claude 4.6 together. What took defense contractors 18 months now takes a dev with a laptop and two API keys. AI isn't replacing devs. It's giving solo builders the firepower of entire teams.
OPENCLAW JUST TURNED DISCORD INTO A 24/7 AI CALL C...
OPENCLAW JUST TURNED DISCORD INTO A 24/7 AI CALL CENTER 🤯 GEMINI 3.1 BUILT IN. IT READS SCREENSHOTS. UNDERSTANDS THE ISSUE. JOINS VOICE CHANNELS. TALKS LIVE. THIS IS NOT “AI REPLIES.” THIS IS AI DOING THE JOB. https://t.co/7rME0PLJa6
Someone built a real-time Intelligence System with...
Someone built a real-time Intelligence System with Gemini 3.1 and Claude 4.6 https://t.co/UROkxqhHB0
Every AI company built a canvas. We built a worksp...
Every AI company built a canvas. We built a workspace. PREVIEW | CODE | BUILD - three views of what your agents are building. In real-time. Multi-model teams debating architecture before writing a single line of code. Claude Opus 4.6 as your Advocate. Gemini 3 Pro as your Critic. They argue tradeoffs, reach consensus, then build. This is https://t.co/qwOAUCDetm right now.
AI agents just leveled up. Hard. No more dumb bot...
AI agents just leveled up. Hard. No more dumb bots. No more single-model lock-in. No more security red flags. Here’s the new play 👇 → Gemini 3.1 built directly into OpenClaw → AI agents that actually think → Discord voice bots that talk in real time → Multimodal understanding (images + text) → Smarter prompt caching = faster automation → Massive security upgrade (SHA-256) → 100+ fixes for real-world reliability This isn’t a small update. It’s a platform shift. If you’re building AI agents, Automating workflows, Or running bots at scale… You need this. Want the full guide? DM me.
Here it is, my recreation of the film Interstellar...
Here it is, my recreation of the film Interstellar in @threejs, 13 iconic scenes, a cinematic montage It’s live and open source, links below 90% of this was built with Codex, I used Opus 4.6 and Gemini 3.1 Pro for many prototypes of scenes, shaders, models (the ref/ dir) https://t.co/uIQ24OwP1F
R.I.P WEB DESIGNER I BUILT THIS WITH SINGLE PROM...
R.I.P WEB DESIGNER I BUILT THIS WITH SINGLE PROMPT USING GEMINI 3.1 BOOKMARK: https://t.co/KdhOcUz3oH
3.1-flash-preview. #Gemini #Gemini3
3.1-flash-preview. #Gemini #Gemini3 https://t.co/qD21mm8Jfl
#Geminiai #Gemini3 なんかGeminiにも音楽生成機能が追加されました。 センシ...
#Geminiai #Gemini3 なんかGeminiにも音楽生成機能が追加されました。 センシティブ判定の変化はこのアップデートが影響してたのかも。 ……ジャズは無いみたいで残念🥲 https://t.co/9up40QmUM2
. exposed the uncomfortable truth about Google's A...
.@xinranma_ exposed the uncomfortable truth about Google's AI design ecosystem. Google has built world-class individual AI tools. They just haven't connected them. "That's the nature of Google products. Stitch, Google AI Studio, they have so many different experiments or products that they are not very well connected, in my opinion." Here is why this matters for anyone building with AI right now: The Google Stitch to AI Studio workflow she demoed: > Take a screenshot of any existing product (he used Redfin) > Paste it into @stitchbygoogle with business and user goals > Set the creative range to "YOLO" mode for maximum divergence > Generate 3+ wildly different redesign options in seconds > Export the best one directly to Google AI Studio > Turn it into a working interactive prototype - for free The hidden limitations he flagged: > Figma export only works in "fast mode" - not across all Stitch models > The Stitch-to-AI-Studio handoff could be far more streamlined but is not > Google could discontinue any of these experiments without warning (as they have historically done) > The redesign model generates creative ideas but sometimes goes "wild on the design system" - diverging far from what was asked