- The Rundown AI
- Posts
- Runway enters the AI image arena
Runway enters the AI image arena
PLUS: Improve your prompts with this new Anthropic feature
Rowan Cheung
November 26, 2024
Sign Up | Advertise | Podcast | AI University
Welcome, AI enthusiasts.
Runway is best known for its AI video prowess, but the startup just released a new image model that may have even the top rivals looking over their shoulders.
With ‘Frames’ turning visual styles into unbelievable explorable worlds, is this the release that turns Runway into a complete AI visual powerhouse? Let’s get into it…
In today’s AI rundown:
Runway unveils ‘Frames’ image generation model
Anthropic launches universal AI connector system
Improve your prompts with this new Anthropic feature
Luma Labs upgrades Dream Machine
5 new AI tools & 5 new AI jobs
More AI & tech news
Read time: 4 minutes
Friendly reminder: We’re moving to a new email sending address. Make sure to add [email protected] to your contacts to ensure you keep receiving our emails.
LATEST DEVELOPMENTS
RUNWAY
Image source: Runway
The Rundown: Runway just revealed a new AI image model called ‘Frames,’ featuring generations with impressive photorealistic quality and stylistic control through distinct ‘Worlds’ that help users maintain consistent aesthetics.
The details:
The new model operates through specialized "World" environments, offering unique artistic directions like vintage film effects and retro anime aesthetics.
Each World is numbered, hinting at a potential library of thousands of available style options and the ability for users to create their own.
Frames will be rolling out inside Runway’s Gen-3 Alpha platform and API, bringing the stylistic control to image-to-video generations.
The launch comes just days after Runway released a video expansion tool that allows users to resize and generate new scenes around an existing video.
Why it matters: Runway is bringing the heat with an image model that looks on par with top rival image startups. Combining Frames with an already powerful Gen-3 Alpha will make for some insanely realistic video generations — and Runway is starting to look like a complete AI visual powerhouse, not just an AI video generation startup.
TOGETHER WITH EIGHT SLEEP
The Rundown: Elevate your sleep this winter with Eight Sleep's Pod 4 Ultra — a high-tech AI-powered sleep solution that enhances recovery and boosts your overall well-being.
The Pod 4 Ultra helps you:
Fall asleep 44% faster with personalized temperature control
Reduce middle-of-the-night wake time by up to 3 hours monthly
Track comprehensive biometrics without wearables
Upgrade your rest this holiday season — visit eightsleep.com/rundown and use code RUNDOWN for up to $600 off bundled purchases until Dec. 14th.
ANTHROPIC
Image source: Alex Albert (@alexalbert_ on X)
The Rundown: Anthropic just launched the Model Context Protocol (MCP), an open-source standard that enables AI systems to directly connect with various data sources and tools — solving the problem of LLMs integrating with external systems.
The details:
The protocol allows AI assistants to access data across repositories, tools, and dev environments through a unified standard.
Anthropic released pre-built MCP servers for popular tools like Google Drive, Slack, and GitHub, and developers can also build their own connectors.
Claude Enterprise users can now test MCP servers locally to connect AI systems with internal datasets and tools.
Anthropic Head of Claude Relations Alex Albert posted a demo showcasing the MCP, with Sonnet 3.5 connecting to GitHub to create a repo and pull request.
Why it matters: As AI assistants evolve into agentic systems, they need seamless access to multiple tools and data sources. The MCP could eliminate the current headache of building separate connectors for every database, tool, and platform — becoming the infrastructure for truly capable AI agents.
AI TRAINING
The Rundown: Anthropic’s new Prompt Improver transforms basic instructions into optimized prompt templates that generate exactly what you need.
Step-by-step:
Visit the Anthropic’s Console dashboard and select "Improve an existing prompt"”
Write your basic prompt using as placeholders.
Describe specifically what you want to improve about your prompt.
Click improve, open in workbench, customize variables, and run!
Pro tip: Save your optimized prompts as templates to maintain consistency across similar tasks.
PRESENTED BY NUME
The Rundown: Nume is an AI CFO that simplifies financial management for startups with instant insights and alerts — helping you spend less time sifting through spreadsheets and more time growing your business.
With Nume, you’ll benefit from:
Beautiful financial reports delivered in under a minute
Instant answers to any financial question
Real-time alerts for important financial events via Slack or email
Register now for early access and never worry about finances again.
LUMA LABS
Image source: Luma AI
The Rundown: Luma AI just released a major upgrade to its Dream Machine platform, introducing a new Photon image generation model alongside a revamped UI featuring new creative control and faster generation speeds.
The details:
The new Photon model claims to be 800% faster than rivals while delivering higher quality outputs and better text generation with more natural prompting.
Dream Machine can now generate consistent characters from a single reference image and maintain them across both images and videos.
The platform also added new camera controls, style transfer, and Brainstorm for creative exploration, moving away from complex prompt engineering.
Dream Machine has four subscription tiers (including a free tier) starting at $9.99/mo, with a $99.99/mo enterprise option for larger teams.
Why it matters: AI image and video models are taking the next step up — but this time, they’re integrated together with a wealth of features to unlock new creative workflows. Luma’s emphasis on natural language and brainstorming is particularly interesting, positioning Dream Machine as a creative partner instead of another AI tool.
NEW TOOLS & JOBS
⚡️ Sparkbase - AI sales agent that combines B2B data with real-time web signals to book sales calls on autopilot
🧠 TwinMind - AI sidebar that listens, sees tabs, and proactively helps users
🤝 Socap AI - AI networking copilot for entrepreneurs
⚒️ Toolhouse - Cloud infrastructure to equip LLMs with actions and knowledge with just three lines of code
⚙️ Lune AI - Community-driven marketplace of individual expert LLMs created on technical topics that outperform standalone AI models
📱 The Rundown - Social Media Manager
💻 Meta - Software Engineer, PyTorch Domains
🎨 Captions - Graphics Software Engineer
📊 Databricks - Collections Lead
🛠️ Perplexity AI - Engineering Manager, SF
QUICK HITS
Free event: Human-AI Collaboration. Learn how to unlock true synergy with AI & what challenges arise in the switch from tool to team member. RSVP for free.*
NVIDIA showcased Fugatto, a 2.5B parameter AI sound model that can generate and transform any combination of music, voices, and audio effects using text prompts and existing audio inputs.
Researchers used AI and drone technology to discover 303 previously unknown Nazca Lines in Peru’s desert, doubling the number of known figures and providing new knowledge of sacred spaces and pilgrimage routes.
U.S. Senator Peter Welch introduced the TRAIN Act, enabling copyright holders to subpoena AI companies' training records when they suspect their work was used without permission to develop AI models.
Perplexity announced a new partnership with Quartr, which will bring the platform AI-powered live earnings call analysis, summaries, and qualitative financial research.
Intuit launched new AI features for its QuickBooks platform, including automated invoice generation, expense categorization, and plans for AI agents that can perform C-suite executive functions.
*Sponsored listing
THAT’S A WRAP
That's it for today!Before you go we’d love to know what you thought of today's newsletter to help us improve The Rundown experience for you. |
See you soon,
Reply