What coding llm is the best?

• supercharger

  13 346 6.6 Python

  Supercharge Open-Source AI Models

I have personally had the best success with a fine tuning Georgia Tech did of Starcoder with an instruction dataset (https://huggingface.co/GeorgiaTechResearchInstitute/starcoder-gpteacher-code-instruct) being run in Supercharger (https://github.com/catid/supercharger). Starcoder itself isn't instruction tuned, and I have found to be very fiddly with prompts. Starcoder I feel takes it to the next level with iterative coding. Supercharger has the model build unit tests, and then uses the unit test to score the code it generated, debug/improve the code based off of the unit test quality score, and then run it all in a loop until it reaches a minimum quality score.

• GPTQ-for-SantaCoder

  5 54 8.2 Python

  4 bits quantization of SantaCoder using GPTQ

This is on my list of projects to explore but haven't made it here yet: https://github.com/mayank31398/GPTQ-for-SantaCoder

• InfluxDB

  www.influxdata.com featured

  Power Real-Time Data Analytics at Scale. Get real-time insights from all types of time series data with InfluxDB. Ingest, query, and analyze billions of data points in real-time with unbounded cardinality.

  

• starcoder.cpp

  3 437 4.8 C

  C++ implementation for 💫StarCoder

  

I'm talking about https://github.com/bigcode-project/starcoder.cpp/tree/main specifically. It is just a command line utility which is why I was wondering if anyone knows of a better way of running it.

Suggest a related project