No it's not very good. But when you run out of Claude tokens it's perfectly fine for small stuff.

Cursor's inline autocomplete is very good though, much better than anything I could reproduce in Zed with various 3rd party "edit" LLMs (although checking google, they announced a new model since I tried it https://zed.dev/blog/zeta2)

I ran parallel prompt with composer 2 and gpt5.3 codex. Composer did slightly better, in terms of variable naming and extra tweaks to loosely related files to keep the codeb consistent.