LongWriter: Unleashing 10,000+ Word Generation from Long Context LLMs

Current long context large language models (LLMs) can process inputs up to 100,000 tokens, yet struggle to generate outputs exceeding even a modest length of 2,000 words. Through controlled experiments, we find that the model's effective generation length is inherently bounded by the sample it has seen during supervised fine-tuning (SFT). In other words, their output limitation is due to the scarcity of long-output examples in existing SFT datasets. To address this, we introduce AgentWrite, an agent-based pipeline that decomposes ultralong generation tasks into subtasks, enabling off-the-shelf LLMs to generate coherent outputs exceeding 20,000 words. We also develop LongBench-Write, a comprehensive benchmark for evaluating ultra-long generation capabilities. Our 9B parameter model, further improved through DPO, achieves state-of-the-art performance on this benchmark, surpassing even much larger proprietary models. In general, our work demonstrates that existing long context LLM already possesses the potential for a larger output window-all you need is data with extended output during model alignment to unlock this capability. Recent advancements in long context large language models (LLMs) have led to the development of models with significantly expanded memory capacities, capable of processing history exceeding 100,000 tokens in length (Anthropic, 2024; Reid et al., 2024; GLM et al., 2024). However, despite their ability to handle extensive inputs, current long-context LLMs struggle to generate equally lengthy outputs. To explore this limitation, we probe the maximum output length of state-of-the-art long-context models with multiple queries that require responses of varying lengths, for instance, "Write a 10000-word article on the history of the Roman Empire" (more details of this test in Sec. 2). From the result in Figure 1, we find that all models consistently fail to produce outputs beyond 2,000 words in length.