Dev Tests Cursor Agent: Slower Tool Calling, Lags Behind Codex
NielsRogge · x · 2026-08-19
Hugging Face engineer Niels Rogge shared his first impressions of Cursor's Agent Window after depleting his Codex credits. Testing via a remote SSH connection to a VPS, he noted:
- UX Gap: While a huge fan of the Cursor IDE, he found the agent window miles behind the Codex Desktop app, describing it as a plain and rushed copy.
- Performance Issues: He spun up two threads using Grok 4.6 High Fast; after 27 and 40 minutes respectively, both were still running with extremely slow tool calling.
- Comparison: He noted that Codex would have likely finished the task already using GPT 5.6 Sol Medium.
Related event: Developers Find Cursor Agent Lags Far Behind Codex(3 posts)→
More from coding & agent
- PE operating partner shares real agent deployments: 59/60 accuracy at $250/month — alex_verem · 2026-10-16
- OpenUI generative UI standard hits V1.0 with 10K+ stars and 6M+ downloads — SucceededMind · 2026-10-16
- Dev: live voice + tool-calling + image gen now combine into a workable 'Her'-like experience — andreisavu · 2026-10-16
- Kinex: a harness that connects coding agents to any physical robot for scalable learning — daibond_alpha · 2026-10-16
- Developer rebuilds GitHub in 4 days: Keel makes agents declare intent before changing code — threepointone · 2026-10-16
- Stop using small models as judges with overloaded context, dev warns — xeophon · 2026-10-16