Tried Gemini 2 weeks ago to see where it's at, with gemini-cli.
Failed to use tools, failed to follow instructions, and then went into deranged loop mode.
Essentially, it's where it was 1.5 years ago when I tried it the last time.
It's honestly unbelievable how Google managed to fail so miserably at this.
Their harness might be behind
I think failures that I observed with gemini are unrelated to the harness. Because the same failures happened with third party harnesses too.
It’s great on AI Studio. Harness issues, I agree.