via twitter, Feb 2022 from Nov 2021
   [1]https://github.com/VHellendoorn/Code-LMs
   [2]https://arxiv.org/abs/2202.13169
   We release a new model, PolyCoder, with 2.7B parameters based on the
   GPT-2 architecture, which was trained on 249GB of code across 12
   programming languages on a single machine. In the C programming
   language, PolyCoder outperforms all models including Codex.

References

   1. https://github.com/VHellendoorn/Code-LMs
   2. https://arxiv.org/abs/2202.13169