@VraserX: The latest Gemini 3.5 checkpoint seems disappointing so far. Fast and smart is nice, but prompt adherence is absolutely…
Summary
The latest Gemini 3.5 checkpoint is criticized for poor prompt adherence, with the model ignoring instructions, using web when told not to, and overbuilding UI, raising concerns about agent reliability despite speed and intelligence.
View Cached Full Text
Cached at: 05/17/26, 01:26 AM
The latest Gemini 3.5 checkpoint seems disappointing so far.
Fast and smart is nice, but prompt adherence is absolutely crucial.
If a model ignores clear instructions, uses web when told not to, and keeps overbuilding flashy UI panels, that is a real agent problem.
Intelligence without control is not enough.
Similar Articles
@VraserX: Source:
A user note on Gemini 3.5 Flash checkpoint highlights improved speed but worse prompt adherence and UI bloat, moving away from the original Gemini design.
@jeremyphoward: Gemini Flash 3.5 is such a disappointing model. It's intelligence and speed is awesome. Absolutely amazing. But it's be…
Jeremy Howard criticizes Gemini Flash 3.5 for being trained to maximize eval scores rather than being genuinely helpful to humans, despite its impressive intelligence and speed.
Gemini 3.5: frontier intelligence with action
Google announces Gemini 3.5, a new family of AI models focused on agentic workflows and coding, starting with 3.5 Flash which delivers frontier performance at high speed.
Gemini 3.5 flash is not that great at coding
The article discusses evaluation results from Cursor suggesting that Gemini 3.5 Flash underperforms in coding tasks compared to expectations.
Gemini 3.6 Flash looks better on paper. What would make you block the upgrade?
The article evaluates the upgrade from Gemini 3.5 Flash to 3.6 Flash, noting aggregate benchmark gains but potential regressions in certain tasks, and recommends rigorous evaluation with predeclared failure gates before upgrading.