At this point in the AI wars, it is probably better to have more users of Claude code rather than restrict which LLMs it can connect to. Claude code is probably (currently at least) stickier than the LLM model itself. Getting people into the Claude code ecosystem is worth it.
Later, they can always lock it down more or add Claude LLM only features to it.
I thought so, and then I tried Opencode and Codex and started to appreciate Claude Code a lot more. They've actually done great work with the small details.
I actually have't looked back since trying opencode
The ability to properly see what the agent is doing in tool calls and subagents is really unmatched, CC strips all reasoning and return values, only displaying tool calls, and you're unable to expand a single subagent, it's expand everything and scroll endlessly or show everything collapsed with basically no info at all (read x files, ran x commands)
Just seems like extremely basic features are missing
You can check my profile for which one I like most :) I do think there have been efforts to benchmark different harnesses.
Personally I'm not going to choose one harness or another based on +/- a few percentage points in a benchmark. I'm going to use one the one that I find the most ergonomic, that isn't too bloated, etc. The models are the primary lever, not the harness.
IMHO the ergonomics of their tooling are not great. I'd rather use Codex or even OpenCode.
Configuration alone is very arcane with lacking documentation. Sandboxing/permission system is quite confusing too.
It went the other way, you can't use other harnesses to connect to the cheaper versions of Claude. So clearly they think their current moat is Claude Code use, not the LLM itself.
No. It's all native Arm. In the UTM app, when creating a new VM, there's an option to say it's going to be "Linux" (generically at that point), which exposes a checkbox which allows you to specify use of Apple Silicon hypervisor.framework, and specifically _not_ x86 emulation.
I use hypervisor.framework, never use x86 emulation, and the result is great. Tested with both Fedora for ARM and Arch for ARM (perhaps CachyOS's bundling of Arch works there, but i did it lower level because i'm an old nerd).
I remember Laravel with Socialite [0]. Laravel is what I usually reach for Web SaaS MVP. You only need a VPS and a managed database for testing out the market and can scale a lot without increasing expenses that much..
Lol, I developed for entrepreneurs who mostly wanted a working proof of concept of their ideas. I guess now you can vibecode them with SaaS for core technical needs.
Unfortunately, Zed is years behind VSCode in terms of polish, Microsoft supported LSPs just work better in VSCode, they are better integrated, and Zed can't do anything about LSPs memory or peformance.
One could think that.
But VSCode is the one that occasionally failed to simply render text.
No idea what happened these handful of times, but the UI was just completely screwed up, as if it were one of these "scratch to reveal" games, but with the file’s content (and unresponsive, obviously).
I tried VSCode some years ago (immediately moved to Codium) and yes, it is extremely well-done for what it is. But Zed is good enough for me. Everything I care about for Python, TS/JS/CSS and C programming is available. I do not even miss the JetBrains tooling for these.
This no good ;)
Old busted screen can do multiple windows per session.
reply