It was so good when it was possible to use Claude subscription on OpenCode.
Nowadays I went from Claude 20x to 5x and been using the GLM model on OpenCode... No regrets.
Nowadays I went from Claude 20x to 5x and been using the GLM model on OpenCode... No regrets.
But actually their own codex harness is quite decent on its own and doesn't have the quality issues or bloat that Claude Code does. Native Rust and open source. And in fact I've got a configuration here to point it at GLM which I also use (via Neuralwatt subscription) in addition to OpenAI's sub.
I do not like opencode's philosophy on the clipboard, it tries to be too clever.
me: Can you implement the next thing
OpenCode+Codex: Yep I'll do that next. <does nothing and returns to prompt>
me: Well?
OpenCode+Codex: <starts implementing>
me: Looks good, let's fix this one issue.
OpenCode+Codex: Sure let's do that. <does nothing and returns to prompt>
me: <bangs head against wall>
--
I've found the codex cli to be much better in this regard, it doesn't nearly derp out so much at higher token counts.
Opus is still my favourite model (I've found 4.6 specifically gives me the best results in OpenCode), but with all the shenanigans Anthropic is pulling, Codex is a close enough substitute.
And yeah, I am supplementing with GLM 5.2 and have actually found it quite complimentary.
One of Codex's weaknesses is "excessive staging" -- basically it's quite cautious and pushes a very incremental approach. This is good for working in an established codebase (which most work is anyways). But for yeeting new projects, Claude always shone better for me (though it often left a mess of race conditions and unhandled negative cases that I had to clean up by hand or with codex)
GLM actually does pretty well in this regard, with the right prompting. It's more "creative" than GPT.