OpenCoder: The Open Cookbook for Top-Tier Code Large Language Models

Code LLMs have been widely used in various domains, including code generation, logical reasoning, and agent systems. However, openaccess code LLMs mostly only release weights, lacking key features such as reproducible data pipelines and transparent training protocols, which are crucial for advancing deeper and more reliable investigations. To address the gap, we introduce OpenCoder, a top-tier code LLM that not only achieves performance comparable to industrial leading models but also serves as an "open cookbook" for the research community. Unlike most prior efforts, we release not only model weights and inference code, but also the reproducible training data, complete data processing pipeline, rigorous experimental ablation results, and detailed training protocols for open scientific research. Our work identifies the key ingredients for building a top-tier code LLM are: language-specific filtering rules, file-level deduplication , highquality synthetic data and two-stage supervised fine-tuning strategy. By offering high level of openness, we aim to broaden access to all aspects of a top-tier code LLM, with Open-Coder serving as both a powerful model and an open foundation to accelerate research, enabling reproducible advancements in code intelligence. The released resource is available at https://opencoder-llm.github.io.