DeepSeek R1 Challenges Frontier Models

Open-weight reasoning model matches GPT-4 class performance at a fraction of the cost.

release6 min read
DeepSeek has released R1, their latest reasoning-focused model...