13 repos
Large language models fine-tuned and optimized for code generation and understanding tasks, built on transformer architectures. The cluster centers on the DeepSeek-Coder family of models across multiple scales (1.3B to 6.7B+ parameters), along with supporting infrastructure for text generation, model serving, and deployment compatibility. These repositories provide both base and instruction-tuned model variants, leveraging frameworks like transformers and text-generation-inference for practical application.