Model Releases
We now have state-of-the-art pre-trained base models in the 3T-parameter weight class, built entirely in-house at MAI. Post-training results…
We now have state-of-the-art pre-trained base models in the 3T-parameter weight class, built entirely in-house at MAI. Post-training results coming soon -- stay tuned! Lots of details in our technical
We now have state-of-the-art pre-trained base models in the 3T-parameter weight class, built entirely in-house at MAI. Post-training results coming soon -- stay tuned! Lots of details in our technical blog post. https://microsoft.ai/news/frontier-pre-training-at-microsoft-ai/
Related
- Block-Level Weight-Space Structure Persists Under Post-Training: An Empirical Study Across LLM Families
- Painless Activation Steering: An Automated, Lightweight Approach for Post-Training Large Language Models
- LoopQuant: Accurate and Efficient Post-Training Quantization for Looped Transformers
- GLM-5.2 is the open-source Claude moment. The demand we’re seeing at Databricks is astonishing. The world is going to see massive adoption o…
Source: Jeremy Howard (X) | 2026-10-15