2024/

📄 2024/06/02/main-swe-bench-html.html 1w HTML · MD

📄 LLMs are bad at returning code in JSON 1,248w HTML · MD

📄 2024/05/22/swe-bench-lite-html.html 1w HTML · MD

📄 Linting code for LLMs with tree-sitter 732w HTML · MD

📄 Coding with Llama 3.1, new DeepSeek Coder & Mistral Large 377w HTML · MD

📄 Claude 3 beats GPT-4 on Aider’s code editing benchmark 623w HTML · MD

📄 Sonnet is the opposite of lazy 651w HTML · MD

📄 Aider has written 7% of its own code (outdated, now 70%) 331w HTML · MD

📄 Drawing graphs with aider, GPT-4o and matplotlib 1,208w HTML · MD

📄 GPT-4 Turbo with Vision is a step backwards for coding 348w HTML · MD

📄 Sonnet seems as good as ever 201w HTML · MD

📄 The January GPT-4 Turbo is lazier than the last version 226w HTML · MD

📄 Aider in your browser 135w HTML · MD

📄 A draft post 21w HTML · MD

Pages