internlm/SWE-Fixer-Editor-72B

TEXT GENERATIONConcurrency Cost:4Model Size:72.7BQuant:FP8Ctx Length:32kPublished:Jan 9, 2025License:mitArchitecture:Transformer0.0K Open Weights Cold

The internlm/SWE-Fixer-Editor-72B model, developed by InternLM, is a 72.7 billion parameter language model fine-tuned from Qwen2.5-7B. It is specifically designed for resolving real-world GitHub issues through a retrieve-then-edit pipeline. This model excels at code editing tasks, making it highly effective for automated software engineering workflows.

Loading preview...

SWE-Fixer-Editor-72B: GitHub Issue Resolution

SWE-Fixer-Editor-72B is a 72.7 billion parameter model developed by InternLM, fine-tuned from Qwen2.5-7B. It is part of the SWE-Fixer project, which focuses on training open-source large language models to effectively and efficiently address real-world GitHub issues.

Key Capabilities

  • GitHub Issue Resolution: Designed to tackle and resolve issues found on GitHub.
  • Retrieve-then-Edit Pipeline: Utilizes a two-component system comprising a code file retriever and a code editor to streamline the issue resolution process.
  • Code Editing: Specialized in editing code to fix identified problems.

Good For

  • Automated software engineering tasks.
  • Developers looking to integrate LLMs into their GitHub workflow for issue fixing.
  • Research and development in automated code repair and software maintenance.