SWE-Llama are variants of the CodeLlama model fine-tuned on software engineering tasks extracted from real-world GitHub issues and pull requests. They were introduced and evaluated on the SWE-bench benchmark.
16 Pulls Updated 7 months ago
Updated 7 months ago
7 months ago
897d459d5e9f · 7.2GB
model
archllama
·
parameters6.74B
·
quantizationQ8_0
7.2GB
Readme
SWE-Llama are variants of the CodeLlama model fine-tuned on software engineering tasks extracted from real-world GitHub issues and pull requests. They were introduced and evaluated on the SWE-bench benchmark.
References: - https://huggingface.co/princeton-nlp/SWE-Llama-7b - https://www.swebench.com/