Forem

Refact AI

Open-source AI code assistant for VS Code and JetBrains

Write better code with a powerful autocompletion fine-tuned on your codebase, ask questions in AI chat aware of your codebase.

Long-term memory in AI programming: Why your team needs an Agent that doesn't forget
Cover image for Long-term memory in AI programming: Why your team needs an Agent that doesn't forget

Long-term memory in AI programming: Why your team needs an Agent that doesn't forget

Comments
6 min read
SWE-bench Multimodal is the benchmark that JavaScript devs might explore
Cover image for SWE-bench Multimodal is the benchmark that JavaScript devs might explore

SWE-bench Multimodal is the benchmark that JavaScript devs might explore

Comments
6 min read
How we built the open-source SOTA AI Agent on SWE-bench Verified
Cover image for How we built the open-source SOTA AI Agent on SWE-bench Verified

How we built the open-source SOTA AI Agent on SWE-bench Verified

1
Comments
9 min read
#1 on SWE-bench lite, achieved fully autonomously by open-source Refact.ai Agent
Cover image for #1 on SWE-bench lite, achieved fully autonomously by open-source Refact.ai Agent

#1 on SWE-bench lite, achieved fully autonomously by open-source Refact.ai Agent

Comments
6 min read
Top-10 Tips for Conscious Vibe Coding
Cover image for Top-10 Tips for Conscious Vibe Coding

Top-10 Tips for Conscious Vibe Coding

9
Comments
8 min read
Refact.ai Agent scores highest on Aider's Polyglot Benchmark: 93.3% with Thinking Mode, 92.9% without
Cover image for Refact.ai Agent scores highest on Aider's Polyglot Benchmark: 93.3% with Thinking Mode, 92.9% without

Refact.ai Agent scores highest on Aider's Polyglot Benchmark: 93.3% with Thinking Mode, 92.9% without

11
Comments 1
6 min read
Our AI Agent + 3.7 Sonnet ranked #1 on Aider’s polyglot bench — a 76.4% score
Cover image for Our AI Agent + 3.7 Sonnet ranked #1 on Aider’s polyglot bench — a 76.4% score

Our AI Agent + 3.7 Sonnet ranked #1 on Aider’s polyglot bench — a 76.4% score

12
Comments
4 min read
Developing AI Agent: SWE bench critics, speed-smart trade-off, and how to make the agent useful
Cover image for Developing AI Agent: SWE bench critics, speed-smart trade-off, and how to make the agent useful

Developing AI Agent: SWE bench critics, speed-smart trade-off, and how to make the agent useful

2
Comments
6 min read
loading...