Ai2's new Olmo 3.1 extends reinforcement learning training for stronger reasoning benchmarks
PositiveTechnology
- The Allen Institute for AI (Ai2) has launched Olmo 3.1, an advanced iteration of its Olmo model family, which enhances reinforcement learning training to improve reasoning benchmarks. This update includes two optimized versions, Olmo 3.1 Think 32B for advanced research and Olmo 3.1 Instruct 32B for instruction-following tasks, alongside a programming-focused model, Olmo 3-Base.
- This development signifies Ai2's commitment to pushing the boundaries of AI capabilities, particularly in efficiency, transparency, and control, which are crucial for enterprise applications. The extended reinforcement learning training schedule aims to bolster the model's performance in complex reasoning tasks.
- The release of Olmo 3.1 aligns with a growing trend in AI towards models that prioritize customization and transparency, as seen in competing models like Qwen and Llama. This reflects a broader industry shift towards enhancing reasoning capabilities and coding skills in AI, which are essential for meeting the increasing demands of diverse applications in technology.
— via World Pulse Now AI Editorial System