OpenAI’s GPT-4.5 is here

Good morning, AI enthusiasts. OpenAI just dropped its largest model yet — but as an alternative of crushing benchmarks, GPT-4.5 is doubling down on *feeling more human*.

With Sam Altman calling it “the primary model that looks like talking to a thoughtful person,” will users embrace this emotional evolution or scoff at the shortage of raw performance gains they’ve come to expect?

In today’s AI rundown:

• OpenAI’s GPT-4.5 with emotional intelligence

• Tencent’s latest ‘fast-thinking’ model

• Constructing a sensible calendar management system

• Ideogram eyeing speed boost with latest model

• 4 latest AI tools & 4 job opportunities

LATEST DEVELOPMENTS

OPENAI

🔥 OpenAI’s GPT-4.5 with emotional intelligence

The Rundown: OpenAI just released GPT-4.5 (code-named Orion), the corporate’s largest model to this point — which uses unsupervised learning as an alternative of reasoning to attain deeper world knowledge and improved emotional intelligence.

The main points:

• OpenAI says GPT 4.5 delivers a more natural conversational experience, with an improved understanding of human intent and greater emotional intelligence.

• The model hallucinates less and delivers more accurate answers than previous versions, with testers liking it for professional tasks, creative work, and on a regular basis queries.

• It’s not a step up from previous models on math or science but does surpass o3-mini and o1 on SWE-Lancer, OpenAI’s latest freelance coding task benchmark.

• Only Pro users and developers on paid plans can access GPT-4.5 immediately, with Plus and Team users gaining access next week.

• Notably, the API price of the model has been kept shockingly high at $75/$150 per million input/output tokens. For reference, GPT-4o costs just $2.50/$10.

Why it matters: While the benchmarks and pricing might leave some disillusioned, 4.5 looks like more of a ‘vibe’ personality upgrade than a significant step up. With high costs and fewer improvements than users have come to expect, this may also be the last stop each practically and acceleration-wise in non-reasoning model development.

TOGETHER WITH TURING

🎤 NeurIPS 2024, decoded

The Rundown: What happened at NeurIPS 2024? This recap breaks down the largest AI advancements from leaders at Oracle, IBM, and Turing so you may stay ahead without the noise.

Gain insights on:

• Multi-agent reasoning: How AI systems are learning to collaborate

• Real-time adaptation: Smarter, more responsive models in motion

• Frontier scaling: Pushing the boundaries of coaching and optimization

Get the complete recap here.

TENCENT

💨 Tencent’s latest ‘fast-thinking’ model

The Rundown: Chinese giant Tencent just released Hunyuan Turbo S, a brand new ‘fast-thinking’ AI designed for immediate responses quite than deep reasoning — achieving 2x the speed while matching the performance of leading models on key benchmarks.

The main points:

• Turbo S matches models like DeepSeek V3, GPT-4o, and three.5 Sonnet across knowledge, mathematics, and reasoning despite a give attention to speed.

• Tencent has significantly lowered the value of the brand new model, making it a fraction of the price of the previous generation.

• The corporate can also be preparing to launch a complementary T1 reasoning model with “deep considering,” positioning the 2 models for various use cases.

• The discharge comes amid increasing AI competition from China, with DeepSeek nearing a brand new launch and Alibaba debuting QwQ-Max for reasoning this week.

Why it matters: It wasn’t way back that reasoning models were the brand new shiny toy, and now now we have a ‘fast-thinking’ vs. ‘slow-thinking’ divide. With DeepSeek’s R1 shining a large global highlight on Chinese AI, rival labs are quickly rushing to one-up the industry darling — and U.S. chip restrictions don’t appear to be slowing anything down.

AI TRAINING

📅 Constructing a sensible calendar management system

The Rundown: On this tutorial, you’ll learn learn how to create a customized AI calendar assistant that may check your schedule and create events using n8n’s workflow automation and AI capabilities.

1. Visit n8n and create a brand new workflow.

2. Add a chat trigger as your entry point.

3. Arrange an AI Agent with calendar tools (Get Events and Create Event) using AI expressions, similar to .

4. Add system instructions for the AI agent node to guide the AI’s behavior.

Pro tip: Test your assistant with natural language prompts like “Discover a one-hour slot tomorrow in my calendar and schedule a research meeting with Alex.”

Take a look at our recent workshop showing learn how to create AI agents with n8n here.

PRESENTED BY AUGMENT CODE

🤖 Where pro engineers meet real AI

The Rundown: Augment Code isn’t your average AI assistant — it’s purpose-built for serious software engineers handling massive codebases. Backed by Eric Schmidt and trusted by top teams like Webflow, Lemonade, and DDN, Augment helps take productivity to the following level.

With Augment Code, you may:

• Immediately integrate into VS Code, JetBrains, or Vim to remain within the zone

• Leverage full-codebase context in under 200ms to maintain your workflow fast

• Handle tens of hundreds of files or thousands and thousands of lines effortlessly

Install today and construct with Augment’s free plan. 

IDEOGRAM

🖼️ Ideogram eyeing speed boost with latest model

The Rundown: Ideogram launched its 2a model, a significant update to the text-to-image platform that significantly reduces generation time and value while maintaining high-quality outputs—with optimizations for graphic design and photorealistic generations.

The main points:

• 2a generates image outputs in only 10 seconds, with an excellent faster ‘2a Turbo’ option delivering results at twice the speed.

• The brand new model excels at graphic design and text generation, with the flexibility to create content like homepages, movie posters, and advertisements.

• It’s also optimized for photorealism and is priced at 50% lower than Ideogram 2.0 for each API and web use.

• Users can access it now via Ideogram’s web platform, API, or through applications like Freepik, Poe, and Gamma.

Why it matters: While many image-generation models still struggle with basic spelling, Ideogram’s latest release is taking graphic design and text generation to completely mindblowing latest levels. With these latest abilities, one other complete step change in creative workflows is about to be unlocked.

QUICK HITS

🛠️ Trending AI Tools

• 💬 Scribe – ElevenLabs’ latest SOTA speech-to-text model

• 🪨 Granite 3.2 – IBM’s compact open models for enterprise use

• 🗣️ Octave TTS – Generate AI voices with emotional delivery

• 🧑‍🔬 Deep Review – AI co-scientist for literature reviews

💼 AI Job Opportunities

• 🏗️ Grammarly – Software Engineer, Android

• 🔧 Scale AI – Field Engineering Tech Lead

• 📊 Dataiku – Senior Renewals Manager

• 🤖 Meta – AI Research Scientist (Robotics)

📰 All the things else in AI today

Pika Labs released its latest 2.2 model, featuring upgraded quality, 10-second, 1080p resolution generations, and latest transition and transformation capabilities.

Meta is reportedly planning to release a standalone Meta AI app in Q2 with potential paid subscription options just like OpenAI’s model.

Figure is pushing up its timeline to bring its humanoid robots into the house, starting Alpha testing this yr because of improvements from its recently revealed Helix AI.

Microsoft rolled out latest updates to Copilot, including a dedicated MacOS app, the flexibility to upload PDF or text files, and UI improvements.

Meta introduced Aria Gen 2 glasses with advanced sensors, on-device AI processing, and an all-day battery for research in machine perception, contextual AI, and robotics.

You Labs unveiled ARI, a research agent able to analyzing as much as 400 sources and generating skilled reports with charts, citations, and visuals in under 5 minutes.

COMMUNITY

🎥 Join our next live workshop

Join our workshop today at 4 PM EST to learn learn how to take full advantage of Claude 3.7 Sonnet’s frontier reasoning capabilities, together with advanced prompting strategies with Dr. Alvaro Cintas, The Rundown’s AI professor.

RSVP here. Not a member yet? Join The Rundown University on a 14-day free trial.

🤝 Share The Rundown, get rewards

We’ll all the time keep this text 100% free. To support our work, consider sharing The Rundown with your mates, and we’ll send you more free goodies.

See you soon,

Rowan, Joey, Zach, Alvaro, and Jason—The Rundown’s editorial team