Deepseek Coder 6.7B Instruct
DeepSeek builds the Deepseek Coder 6.7B Instruct, a chat model exceling at code completion, generation, and review tasks, thanks to its pre-training on a large project-level code corpus with a 16,384 token context window. This model is notably flexible, offered in various sizes to suit different user requirements, and achieves state-of-the-art performance among open-source code models on multiple benchmarks, including HumanEval and MultiPL-E. With a massive 2T token training dataset, it demonstrates superior capabilities in coding tasks.
Input
Output
Context
16K
Max Output
-
Parameters
6.7B
Technical Specifications
Input Modalities
Output Modalities
Capabilities
Resources & Links
Estimated GPU Requirements for Deepseek Coder 6.7B Instruct
Estimates based on INT8 quantization. Actual requirements vary by framework and configuration.
Related Tools
Data sourced from official provider APIs and documentation
Last updated: Jun 24, 2026
Start building with the right model.
Automatically route workloads to the right model for every task, every time.