SmolLM3: smol, multilingual, long-context reasoner

SmolLM3 is a new multilingual language model designed for long-context reasoning, featuring an architecture optimized for efficiency with a model size of 1.3 billion parameters. It achieves state-of-the-art performance on several long-context benchmarks and supports multiple languages, making it suitable for applications requiring extensive contextual understanding across diverse linguistic inputs. This model provides practitioners with a compact yet powerful tool for tasks that demand both multilingual capabilities and long-range reasoning, enhancing the versatility of LLM deployments.