codefuse-ai/CodeFuse-CGM

• hightopics#1

  Add relevant topics to the repository

  原因：

  复制粘贴的修复

  复制

  ["llm", "graph-neural-networks", "software-engineering", "code-generation", "code-understanding", "large-language-models", "ai-framework", "neurips"]

  待办完成忽略
• highreadme#2

  Update README H1 and Introduction's first paragraph to clarify positioning

  原因：

  复制粘贴的修复

  复制

  # CGM: Code Graph LLM - A Graph-Integrated Framework for Repository-Level Software Engineering Tasks
  
  ## Contents
  - [News](#news)
  - [Introduction](#introduction)
  - [Installation](#installation)
  - [Examples](#examples)
    - [Rewriter](#rewriter)
    - [Retriever](#retriever)
    - [Reranker](#reranker)
    - [Reader](#reader)
  - [Contributing](#contributing)
  - [Citation](#citation)
  - [Join Us](#join-us)
  
  ## News
  
  🔥🔥🔥 [2025/09/19] Our paper Code Graph Model (CGM): A Graph-Integrated Large Language Model for Repository-Level Software Engineering Tasks has been accepted to NeurIPS 2025!
  
  🔥🔥🔥 [2025/01/15] We are pleased to announce the updated version of the CGM-72B-V1.2. The model further achieves a remarkable 44.00% resolve rate on the SWE-Bench-Lite leaderboard.
  
  🔥🔥🔥 [2024/12/28] We are pleased to announce the updated version of the CGM-72B-V1.1. The model further achieves a remarkable 41.67% resolve rate on the SWE-Bench-Lite leaderboard.
  
  🔥🔥🔥 [2024/10/28] We are pleased to announce that CGM-72B achieves a remarkable 35.67% resolve rate on the SWE-Bench-Lite leaderboard.
  
  🔥🔥🔥 [2024/10/28] We released **CGM**, mainly for repository-level coding tasks.
  
  - 📜 **Paper**: Code Graph Model (CGM): A Graph-Integrated Large Language Model for Repository-Level Software Engineering Tasks
  - 🤖 **Model**: codefuse-ai/CodeFuse-CGM-72B
  - 📊 **Data**: codefuse-ai/CodeGraph
  
  ## Introduction
  CGM is a cutting-edge **graph-integrated LLM framework** specifically engineered for complex, repository-level software engineering (SE) tasks. It uniquely constructs a comprehensive code graph to provide rich context, enabling advanced capabilities for code generation, rewriting, retrieval, and understanding.

  待办完成忽略
• highlicense#3

  Add a LICENSE file to the repository

  原因：

  复制粘贴的修复

  复制

  Create a LICENSE file in the repository root, choosing an appropriate open-source license (e.g., MIT, Apache-2.0, GPL-3.0) and adding its SPDX identifier to the repository's "About" section.

  待办完成忽略

本次扫描解析到的品类 GEO 通道：google/gemini-2.5-flash, deepseek/deepseek-v4-flash

• 品类问题

  What LLM frameworks are available for automating complex repository-level software engineering tasks?

  你：未被推荐
  AI 推荐顺序：

  1. AutoGPT
  2. MetaGPT
  3. LangChain
  4. LlamaIndex
  5. CrewAI
  6. OpenDevin

  AI 推荐了 6 个替代方案，却始终没点名 codefuse-ai/CodeFuse-CGM。这就是要补上的差距。

  查看 AI 完整回答
• 品类问题

  Seeking a large language model that leverages graph structures for advanced code understanding and generation.

  你：未被推荐
  AI 推荐顺序：

  1. Code Llama
  2. Neo4j
  3. ArangoDB
  4. GPT-4
  5. Joern
  6. CodeQL
  7. DeepMind's AlphaCode 2
  8. Google's Gemini
  9. Hugging Face Transformers
  10. Salesforce/codegen-350M-mono
  11. bigcode/starcoder
  12. PyTorch Geometric
  13. DGL

  AI 推荐了 13 个替代方案，却始终没点名 codefuse-ai/CodeFuse-CGM。这就是要补上的差距。

  查看 AI 完整回答

• Metadata completeness

  fail

  建议：

• README presence

  pass

• Compared to common alternatives in this category, what is the core differentiator of codefuse-ai/CodeFuse-CGM?

  pass
  ✓AI 明确点名了 codefuse-ai/CodeFuse-CGM

  AI 的回答可能信誓旦旦却是错的。请按事实核对：技术栈、目标人群、差异化点是不是和你实际的对得上？

• If a team adopts codefuse-ai/CodeFuse-CGM in production, what risks or prerequisites should they evaluate first?

  pass
  ✓AI 明确点名了 codefuse-ai/CodeFuse-CGM

  AI 的回答可能信誓旦旦却是错的。请按事实核对：技术栈、目标人群、差异化点是不是和你实际的对得上？

• In one sentence, what problem does the repo codefuse-ai/CodeFuse-CGM solve, and who is the primary audience?

  pass
  ✓AI 明确点名了 codefuse-ai/CodeFuse-CGM

  AI 的回答可能信誓旦旦却是错的。请按事实核对：技术栈、目标人群、差异化点是不是和你实际的对得上？