LLM

From the evolution of the GPT series, researchers have gradually realized that as long as model parameters, training data, and compute resources are continuously scaled up, the performance of large models improves along a stable and predictable path. This predictability is characterized by Scaling Laws, which provide the theoretical foundation and practical confidence for high-cost pre-training. As model scale, alignment techniques, and inference-time compute co-evolve, the boundaries of AI capabilities are being systematically pushed. Scaling laws are not only the foundation for building next-generation models but also a key methodology for continuously improving model capabilities under compute constraints. ...

As Large Language Models (LLMs) achieve breakthroughs in natural language processing, their applications continue to expand. However, they also exhibit limitations such as knowledge cutoffs, hallucinations, and deficiencies in complex computation and logical reasoning. To address these challenges, Agentic RL, which combines agents with Reinforcement Learning (RL), is emerging as a key research direction. Agentic RL enables LLMs to possess capabilities like autonomous planning, decision-making, tool use, and environmental interaction by creating a closed-loop interaction with the external world (e.g., search engines, code interpreters, databases, browsers) and continuously optimizing through reward signals. In practical applications, it not only understands requirements and plans autonomously but also constantly corrects and optimizes within an execution-feedback loop. ...

In recent years, Large Language Models (LLMs) have achieved revolutionary breakthroughs in fields such as natural language processing, code generation, and even multimodal interaction. However, the powerful capabilities of these models come at the cost of enormous computational and memory overhead, especially during the inference stage. Efficiently deploying and running these models, which have billions or even trillions of parameters, has become a core challenge in scaling LLM technology for real-world applications. ...

DeepSeek AI successively released DeepSeek-V2 (DeepSeek-AI, 2024) and DeepSeek-V3 (DeepSeek-AI, 2024), two powerful Mixture-of-Experts (MoE) language models that significantly optimize training costs and inference efficiency while maintaining state-of-the-art performance. DeepSeek-V2 has a total of 236B parameters, activating 21B per token, while DeepSeek-V3 further expands to 671B total parameters, activating 37B per token. Both support a 128K context length. The core innovations of these two models lie in the adoption of Multi-head Latent Attention (MLA) and the DeepSeekMoE architecture (Dai et al., 2024). MLA drastically reduces GPU memory usage during inference by compressing the Key-Value (KV) cache into low-dimensional latent vectors, improving efficiency. DeepSeekMoE achieves stronger expert specialization capabilities and more economical training costs through fine-grained expert segmentation and shared expert isolation. Building upon V2, DeepSeek-V3 further introduces an Auxiliary-Loss-Free Load Balancing strategy (Wang et al., 2024) and the Multi-Token Prediction (MTP) (Gloeckle et al., 2024) training objective, further enhancing model performance and training efficiency. ...

LLaMA The LLaMA series of open-source models released by Meta AI has become one of the cornerstones of the large language model community, profoundly impacting the advancement of open research and applications. From the pioneering LLaMA released in early 2023, to the significantly improved LLaMA 2 later that year, to derivative models targeting specific domains (like code, safety), and the subsequent new generations LLaMA 3 and LLaMA 4 launched in 2024 and 2025 respectively, Meta has continuously committed to enhancing the performance of open-source models, gradually bringing them closer to state-of-the-art closed-source models. Below, we will introduce the key technical details of each major model in sequence. ...