Shypd.aiThe largest open index of AI toolsOpenCoder-llm is a Coding & Development tool that provides an open and reproducible framework for building top-tier code large language models. It includes model weights, training data, and a complete data processing pipeline.
Trending
About
OpenCoder-llm offers an open and reproducible code LLM family, including 1.5B and 8B base and chat models, supporting both English and Chinese. It is pretrained on 2.5 trillion tokens of raw code and code-related web data, and fine-tuned on over 4.5M high-quality SFT examples. The project provides not only model weights and inference code but also reproducible training data, a complete data processing pipeline, rigorous experimental ablation results, and detailed training protocols. Key features include complete open-source transparency, comprehensive experimental analysis, high-quality synthetic data generation, and exceptional performance across benchmarks, empowering researchers to build and innovate in code AI.
Capabilities
Pricing & Plans
Open Source
Free
FAQs
Trending
