In terms of using experience, I found Grok 4.5 to be way more pleasant to use than GPT 5.6 Sol and Claude 4.8/5. It just gets to the point, and is super fast and concise, no yapping. That's how AI agents should be imo. None of the weird "Claude ipsum" jargon like "load-bearing" and "stale folklore" or GPT 5.6-isms like "focused regression" and "provenance".
I’m not doing any coding with AI, so I’m the odd one out. Mostly use it for research: information retrieval and grokking technical concepts for exam prep. Does that fall within “knowledge work”?
Anywho - I switched to Opus last week and felt torn. It’s displayed somewhat higher competency in some responses, and the artifacts (diagrams) are splendid, but I despise its writing style. Grok is indeed fact/truth oriented, direct, and less personable (which I vastly prefer). Maybe I’ll switch back to Grok.
I don't think it's as significant anymore but there was a point where GPT would write 8 paragraphs to say yes, while Grok would literally just say "Yes." They've converged a bit (Grok now says more, GPT says less) but I'd argue Grok is still more efficient while Claude/GPT are more wordy (and often needlessly)
These differences are prominent even in the same harness (cursor) in my experience. Opus is wordy to the point of exhaustion, GPT is better, Grok is very direct.
