Gemini 2.5 Pro

When Google rolls out new model versions, previously stable models like Gemini 1.5 Pro and 2.0 Flash experience intermittent failures and massive latency spikes (milliseconds to 15+ seconds), with issues persisting for days. Function-calling feature has intermittently failed for 3+ day periods.

The official stable release of Gemini 2.5 Pro is perceived as a significant downgrade from the preview '03-25' version. Developers report the production model as 'lobotomized' with increased hallucinations, worse instruction-following, and excessive verbosity despite explicit directives to be concise.

Gemini models (especially 2.5-Pro) generate code with excessive comments, which is good for learning but poor for production agentic coding. The verbose commenting style makes AI-generated code immediately identifiable and impractical for autonomous agents.