Good day, creatives
In today’s Newsletter:
1️⃣ The AI Race is hotter than ever: Claude Sonnet 3.7 vs OpenAI GPT 4.5
2️⃣ Amazon launches Alexa+ with Claude AI
3️⃣ Alibaba brings more competition to AI Video with its model Wan 2.1
4️⃣ Ideogram 2a has been released, and it’s everywhere
5️⃣ Quick news you may have missed
6️⃣ Quiz: Can you identify AI Art?
7️⃣ Prompt Fight: Noir Drama
8️⃣ Video AI of the week
Let’s go.
The AI Race is hotter than ever
Last week, Grok 3 dominated the headlines.
Then came Claude Sonnet 3.7.
And, of course, OpenAI wasn’t going to sit still, was it?
Claude 3.7 Sonnet takes the lead in the AI race
Anthropic has unveiled Claude 3.7 Sonnet, its most advanced AI yet and the first-ever hybrid reasoning model. This innovation allows Claude to provide either quick responses or deep, structured reasoning, depending on user needs, a new approach in AI.
Key Features & Advancements
Extended Thinking Mode – Users can see Claude’s reasoning process before it answers, adjusting thinking depth.
Claude Code – A command-line tool for searching, editing, testing, and pushing code to GitHub.
A Glimpse into AI’s Future
Anthropic’s roadmap envisions Claude evolving from an assistant (2024) to a collaborator (2025), and eventually a pioneer (2027)—capable of solving problems that would take human teams years. With a $3.5B funding round on the horizon, expect more major updates soon.
Finally, OpenAI releases GPT-4.5. No one is impressed
OpenAI has officially announced GPT-4.5. While it may not dominate benchmarks, early impressions suggest a transformative user experience.
GPT-4.5 excels in creative writing, ideation, and practical advice, while models like o3-mini remain preferable for complex calculations. OpenAI enhanced its performance by leveraging multiple data centers and synthetic data, reducing hallucinations.
Why this is important?
GPT-4.5 has received a mixed response. While some users welcome its release, much of the discussion has centered on its pricing: $75 per million input tokens and $150 per million output tokens. Even OpenAI CEO Sam Altman acknowledged its high cost, calling GPT-4.5 a 'giant, expensive model.
Amazon launches Alexa+ with Claude AI
Amazon is upgrading Alexa with Claude-powered AI in its new Alexa+ assistant. Developed in collaboration with Anthropic, Alexa+ enhances voice interactions with smarter, more natural responses.
Pricing & Benefits
Alexa+ costs $19.99/month but is free for all Prime members. It’s going to be rolling out in the U.S. over the next few weeks, prioritizing Echo Show 8, 10, 15, and 21 users.
Alibaba brings more competition to AI Video with its model Wan 2.1
Alibaba has announced the open-source release of its video and image generating AI model Wan 2.1, making it available on Krea, Nim and Fal. The move follows similar steps by DeepSeek, signaling increased competition in AI development.
Why This Matters
Alibaba’s open-source approach lowers barriers to AI adoption, fosters innovation in AI-generated media, and challenges the current leading players.
Ideogram 2a has been released, and it’s everywhere
Ideogram has launched Ideogram 2a, its fastest and most cost-effective text-to-image model, optimized for graphic design and photography: If you want text on your art, this is your best bet.
Now available on the Ideogram website, API, and partner platforms, it delivers high-quality text-integrated images in just 10 seconds—or 5 seconds with 2.0 Turbo. Pricing has been cut by 50%, making it more accessible than ever. Try it now through partners like Freepik, Fal or Krea AI.
Quick news you may have missed
Luma’s DreamMachine now supports Video-to-Audio, allowing users to generate sound for their AI-created videos. Simply select the new "Audio" button to add sound instantly or customize it with prompts.
Currently in beta and free for all users, this update enhances DreamMachine’s creative workflow by integrating AI-generated visuals and audio.
Kaiber AI launched SuperStudio Pro, which provides a single infinite canvas where you can generate videos and images, create custom models, and build dynamic workflows using tools from multiple AI providers.
ElevenLabs introduces Scribe, the most accurate Speech-to-Text model. Excelling in major languages like English, Spanish, and Italian, it offers smart speaker diarization to tell you who is speaking, timestamp at word level for accurate subtitles, and auto-tagging sound events like audience laughters.
OpenAI is finally making its video generation model Sora, available in the UK and Europe, Switzerland, Norway, Liechtenstein, and Iceland. The company said that ChatGPT Plus and Pro subscribers in these regions will be able to create videos using the model.
Quiz: Can you identify AI Art?
The answer is at the end of the post.
Prompt Fight: Noir Drama
Let’s see how the top AI Art generators compare to each other, with a prompt shared by @LudovicCreator
Prompt: A mysterious femme fatale stepping out of a dimly lit speakeasy portrayed in a Grayscale Cinematic Drama, inspired by classic film noir. Use strong contrasts and dramatic shadows to create a sense of tension and intrigue, with subtle variations of black and white
Which one wins for you?
AI Video of the Week
Abel Art uses AI to tell a simple but powerful story.
Answer to the Quiz:
Alex Ross did the cover for Immortal Hulk #18
Thanks for reading. See you next week!