• https://arxiv.org/abs/2309.12499

  • https://arxiv.org/abs/2310.02003v1

  • Benchmarks

    • https://blog.continue.dev/an-introduction-to-code-llm-benchmarks-for-software-engineers/
  • Design generation

    • The Figma prototype: I build my ideas #8 - 07/19/20

      Untitled

    {
        "navigation": [
            {"icon": "camera"},
            {"title: "Photos"},
            {"icon": "message"}
        ],
        "photos": [
            {"icon": "user"},
            {"frame": "photo"},
            {"icon": "heart"},
            {"icon": "chat bubble"}
        ]
    }
    
  • Code generation

    • Builder codegen
    • React Tailwind code:

    https://twitter.com/gabe_ragland/status/1598068207994429441

    • https://www.patterns.app/blog/2023/01/18/crunchbot-sql-analyst-gpt/ “given this database .. generate query to answer .. now check for these mistakes. Now fix query because db returned error .. now loop”
    • https://twitter.com/thekitze/status/1639724609112096768?s=20
  • Interactive codebase tools

    • https://github.com/BloopAI/bloop: Code search and explanation
    • Copilot X
    • https://www.cursor.so/: IDE, with next edit prediction
    • https://www.codium.ai/: test generation, code search
    • Cody from Sourcegraph (OSS): like an open source Copilot
    • https://github.com/CodedotAl/gpt-code-clippy: abandoned
    • https://github.com/mpoon/gpt-repository-loader: just concat
    • refact.ai: completion, refactoring, chat
    • https://reviewify.io/
    • https://continue.dev/ (Apache): somewhat Cody like
    • LocalPilot: just routes Copilot requests to a local LLM instead of Github; can use mitmproxy instead (HN)
    • Codeium
    • https://github.com/huggingface/llm-vscode completions
  • Coding agents

    • aiden: edit existing codebases
    • ‣ (MIT): generate from scratch. uses tools?
    • https://github.com/AntonOsika/gpt-engineer (MIT): generate from scratch
    • https://github.com/emrgnt-cmplxty/automata (Apache)
    • smol https://github.com/smol-ai/developer(MIT)
    • https://github.com/sweepai/sweep (Elastic): edit existing codebases
    • https://github.com/joshpxyne/gpt-migrate (MIT)
    • Codegen.com
    • https://blog.research.google/2023/05/large-sequence-models-for-software.html?m=1
    • https://blog.research.google/2022/07/ml-enhanced-code-completion-improves.html?m=1
  • Code interpreter

    • https://github.com/microsoft/autogen
    • https://github.com/KillianLucas/open-interpreter
  • Easy end-user inference: llama.cpp, ollama, gpt4all