# Apr 12, 2025

## Google Is Winning on Every AI Front

- **Google DeepMind's Gemini 2.5 Pro** is currently the leading AI model, outperforming competitors like OpenAI and Anthropic across multiple benchmarks, including [LMArena](https://lmarena.ai/?leaderboard) and [Humanity's Last Exam](https://scale.com/leaderboard/humanitys_last_exam).

- **Google's Gemini AI models** will soon be operable in clients' own data centers, enhancing data control and security, with early access slated for the **third quarter** of 2025.

- **AI code assistants frequently hallucinate non-existent package names**, with a staggering **21.7%** of suggestions from open-source models being fictitious, posing significant risks to software supply chains.

- **Python can be compiled into native code**, significantly enhancing its execution speed to match that of Rust, which is crucial for performance-intensive applications like AI and scientific computing.

- **d1** is a novel framework that enhances **reasoning capabilities** in diffusion-based large language models (dLLMs) through a combination of **supervised finetuning** and a unique **critic-free RL algorithm** called diffu-GRPO.

- The team has developed an **AI-native runtime** that can **snapshot-load LLMs** (13B–65B) in **2–5 seconds**, enabling the dynamic execution of **50+ models per GPU** without constant memory residency.

- **Google's decision to allow enterprises to self-host SOTA models** marks a significant shift in the landscape of AI, enhancing **data privacy** and control for businesses, akin to Mistral's existing model.

- **Adding new vocab tokens** to LLMs during instruction-tuning has proven **ineffective**, as models using the base tokenizer show superior validation losses and output quality compared to those with modified tokenizers.