supercharger VS turbopilot

Since I've been on a AI code-helper kick recently. According to the post, Claude 2 now 71.2%, a significant upgrade from 1.3 (56.0%). It isn't specified whether this is pass@1 or pass@10.

For comparison:

* GPT-4 claims 85.4 on HumanEval, in a recent paper https://arxiv.org/pdf/2303.11366.pdf GPT-4 was tested at 80.1 pass@1 and 91 pass@1 using their Reflexion technique. They also include MBPP and Leetcode Hard benchmark comparisons

* WizardCoder, a StarCoder fine-tune is one of the top open models, scoring a 57.3 pass@1, model card here: https://huggingface.co/WizardLM/WizardCoder-15B-V1.0

* The best open model I know of atm is replit-code-instruct-glaive, a replit-code-3b fine tune, which scores a 63.5% pass@1. An independent developer abacaj has reproduced that announcement as part of code-eval, a repo for getting human-eval results: https://github.com/abacaj/code-eval

Those interested in this area may also want to take a look at this repo https://github.com/my-other-github-account/llm-humaneval-ben... that also ranks with Eval+, the CanAiCode Leaderboard https://huggingface.co/spaces/mike-ravkine/can-ai-code-resul... and airate https://github.com/catid/supercharger/tree/main/airate

Also, as with all LLM evals, to be taken with a grain of salt...

Liu, Jiawei, Chunqiu Steven Xia, Yuyao Wang, and Lingming Zhang. “Is Your Code Generated by ChatGPT Really Correct? Rigorous Evaluation of Large Language Models for Code Generation.” arXiv, June 12, 2023. https://doi.org/10.48550/arXiv.2305.01210.

Leverage locally-hosted Large Language Models to write software + unit tests (https://github.com/catid/supercharger)

supercharger Write Software + unit tests for you, based on Baize-30B 8bit, using model parallelism

Added support for 65B LLaMa model to https://github.com/catid/supercharger tonight. It runs faster than Baize 30B (maybe due to lack of adapter) and only slightly slower than Galpaca 30B. Benchmarks here: https://docs.google.com/spreadsheets/d/1TYBNr_UPJ7wCzJThuk5ysje7K1x-_62JhBeXDbmrjA8/edit?usp=sharing

Here's the code that loads it: https://github.com/catid/supercharger/blob/main/server/model_koala.py

fyi https://github.com/ravenscroftj/turbopilot

TurboPilot: self-hosted copilot clone which uses the library behind llama.cpp to run the 6 Billion Parameter Salesforce Codegen model in 4GiB of RAM. (https://github.com/ravenscroftj/turbopilot)

This repo has a potential

Right now https://github.com/ggerganov/llama.cpp is the dominant back-end for querying models, but forks and alternatives like https://github.com/ravenscroftj/turbopilot keep popping up. Increasingly, models submitted to huggingface explicitly note in their READMEs that the model is not compatible with llama.cpp, and that a different back-end must be used.

There's turbopilot. I haven't tried it yet, but it looks promising.

Turbopilot | open source LLM code completion engine and Copilot alternative

This 6B parameters tool (based on README) could be runned with 4 Gb of RAM. https://github.com/ravenscroftj/turbopilot

there is an interesting link https://github.com/ravenscroftj/turbopilot/wiki/Converting-and-Quantizing-The-Models , just wondering if anyone have done this with 16b and put the weights somewhere