Hello Tuners,

Tune Chat has rolled out significant features aimed at enhancing collaboration and efficiency. The update includes expanded integrations, improved model management tools, and real-time file interaction. Meanwhile, Sriram Krishnan, former a16z partner, has been appointed as senior AI policy advisor at the White House, bringing his expertise to shape national AI policy.

In other news, OpenAI’s o3 model is setting new benchmarks in AI scalability, though its high computational demands make it out of reach for everyday users. Additionally, Elon Musk’s xAI raised $6 billion in a Series C round, which will fuel further product development and infrastructure growth.

• Tune Chat Receives Major Update

• The AI Show with Elon and Sriram

• o3: Here’s How Much You Are Spending

• Marketers On Alert for Generative Engine Optimization

Tune Chat Receives Major Update

Tune Chat was released with a significant update designed to enhance user experience. Based on feedback, the platform has new features to help you collaborate and work more efficiently. Join 620,000 users working across different use cases, models, providers, and more!

• Model Management: Simplify workflows with tools to organize, sort, and switch between models.

• API Support: Seamlessly integrate custom models into your workspace.

• Collaboration Features: Share chats and workspaces with team members and organizations.

• Expanded Integrations: Now compatible with OpenAI, Anthropic, OpenRouter, and more.

• Global Search: Locate files and chats across your workspace instantly.

• File Interaction: Organize and engage with your files in real time.

Invite your friends and family to join Tune Chat and enjoy the perks together. For every friend who signs up, you’ll both receive a free month of Pro, valued at $20 each. It’s our way of saying thank you for being part of our journey!

The AI Show with Elon and Sriram

President-elect Donald Trump has tapped Sriram Krishnan, former a16z general partner, as senior policy advisor for AI at the White House’s Office of Science and Technology Policy. Krishnan, a seasoned tech entrepreneur with stints at Microsoft, Twitter, and Facebook, will help steer AI policy across the government. He’ll also work alongside David Sacks, Trump’s new crypto and AI "czar."

Known for his podcast “The Aarthi and Sriram Show” and his close ties with Elon Musk, Krishnan brings a mix of venture capital smarts and a strong stance on keeping the internet open (he’s not a fan of AI models locking down the web). Krishnan's new gig will surely bring some interesting developments with his experience running product teams and time at a16z.

o3: Here’s How Much You Are Spending

OpenAI’s o3 model has set a new benchmark in AI scaling, scoring 88% on the ARC-AGI test and outpacing previous models. However, this progress comes with a hefty price tag, as o3 uses far more computing than earlier versions. The model’s success hints at a faster AI development trajectory in 2025, though it’s likely to remain out of reach for everyday users due to high costs.

While o3 excels at complex tasks, its high computing demands make it impractical for everyday queries. Test-time scaling is being hailed as the next big step in AI, but its hefty resource requirements mean that only deep-pocketed institutions will benefit for now. Despite its flaws, like occasional hallucinations, o3 shows that AI's scaling potential is far from tapped out.

xAI Grabs Pre-Holidays Series C

xAI, Elon Musk’s AI company, has raised $6 billion in a Series C round, bringing its total funding to $12 billion. This round saw investments from firms like Andreessen Horowitz, Blackrock, and Kingdom Holdings, contributing $400 million. xAI is now valued at $45 billion, nearly double its previous valuation. Musk aims to use the funds to accelerate infrastructure, product development, and R&D.

The company’s flagship model, Grok, powers features on X (formerly Twitter), including a chatbot for Premium subscribers. Grok, known for its "rebellious streak," offers a more unfiltered experience than competitors. Despite fierce competition from OpenAI and Anthropic, xAI plans to integrate its models with Tesla and SpaceX technologies, driving significant revenue. As it expands, xAI continues to invest heavily in infrastructure, including a data center in Memphis powered by 100,000 Nvidia GPUs.

Weekly Research Spotlight 🔍

Genesis

The Genesis project, a result of a 24-month collaboration involving over 20 research labs, introduces a groundbreaking generative physics engine. Genesis can create 4D dynamical worlds and is designed for general-purpose robotics and physical AI applications. Developed in pure Python, it is 10-80x faster than existing GPU-accelerated solutions like Isaac Gym and MJX, with a simulation speed of ~430,000 times faster than in real-time. With a transferable real-world impact, it can train a robotic locomotion policy in just 26 seconds on a single RTX4090.

Genesis is fully open-source, available at Genesis on GitHub, and offers a unified simulation framework that integrates advanced physics solvers for unparalleled realism. The platform will soon roll out access to its generative framework, allowing for creating virtual environments, robot policies, character motions, and interactive 3D scenes—paving the way for fully automated data generation in robotics and AI.

LLM Of The Week

QVQ-72B-Preview

QVQ-72B-Preview has shown impressive capabilities across several benchmarks. It scored 70.3% on the Multimodal Massive Multi-task Understanding (MMMU) benchmark, highlighting its strength in multidisciplinary tasks. The model also excelled in MathVision, showcasing advanced mathematical reasoning, and demonstrated enhanced problem-solving on OlympiadBench.

Despite these achievements, the model faces challenges in language mixing, recursive reasoning loops, and visual reasoning. While it outperforms many models, its performance in basic recognition tasks and multi-step visual reasoning is still behind Qwen2-VL-72B.

Best Prompt of the Week 🎨

Today's Goal: Try new things 🧪

Acting as a Career Planning Guide

Prompt: I want you to act as a career planning guide. You will create a structured daily plan specifically designed to help an aspiring manga artist hone their craft and prepare to join a prestigious manga agency. You will identify key strategies for improving essential skills such as storytelling, character design, panel composition, and inking. Additionally, you will outline steps to build a strong portfolio, gain exposure through social media or independent publishing, and develop a consistent work ethic to meet industry standards. My first suggestion request is: "I need help creating a daily activity plan for someone starting their journey as a manga artist and aiming to join a top-tier manga agency."

This Week’s Must-Watch Gem 💎

This Week's Must Read Gem 💎