I've been using both Continue and Cursor on my M1 MacBook Pro (16GB RAM) for a few weeks now, primarily for API integration work and reviewing automation scripts. I kept hearing that Cursor was more resource-intensive, but I wanted to see the actual memory footprint during my typical workflow.
I used Activity Monitor to track both over several days, with similar usage patterns:
- Having 2-3 projects open (Node.js/Postman configs)
- Using the chat for code explanation and generation
- Running inline edits on existing Zapier task automations
Here's what I observed during active development sessions:
**Continue (v0.23.4):**
- Baseline: 180-220 MB
- With chat active + file context: 280-350 MB
- Peak during large file processing: 420 MB
**Cursor (v0.37.2):**
- Baseline: 420-480 MB
- With chat active + file context: 650-800 MB
- Peak during agentic workflow: 1.2 GB
The surprising part wasn't just the higher baseline, but how Cursor's memory usage would "stick" after complex operations. Continue would reliably drop back to its baseline after a few minutes, while Cursor would often maintain 600+ MB even during idle periods.
For my workflow—which involves a lot of switching between documentation, API specs, and existing code—the lower memory footprint of Continue means I can keep other tools open (like my CRM's web app, Postman, and Workato designer) without constant memory pressure warnings.
Has anyone else done similar comparisons? I'm particularly curious if this holds true for larger projects or when working with multiple LLM providers.