Gemini 3
Reddit take
Not enough linked mention summaries yet; the source excerpts below are the evidence.
Pros
- Gemini 3 is great, but it seems that right now, only Code offers it, not VS26.
- Gemini 3 is better tuned to natural language and collaborative communication. I have already seen better results from this as opposed to concise direction. The fact that it will tell me that it isn't sure rather than coming up with a fa...
- But the real power of the 2025 models (like Gemini 3 and GPT-5.2) isn't in their "knowledge", it's in their ability to perform **System 2 thinking**
- So I brought in the best prompt engineer I know: Gemini. Using the map context, I ask Gemini 3 to generate a detailed structured prompt as JSON: camera settings, composition, lighting, quality, everything.
- With the release of Gemini 3 and Opus 4.5, I needed a prompt to generate landing pages quickly to test frontend capabilities of these models.
Cons
- Too many people are approaching Gemini 3 as a hammer still, and they keep smacking it's head against a nail and questioning why the nail drives through the head and not the wood.
- caveat: I haven't tried Gemini 3 for poetic content; so may be good or bad can't vouch
- Gemeni is meh.
- Gemini 3 is showing improvement, but in my opinion, it’s still not there yet.
- However, I'd avoid the GPT-5 family and Gemini 3 entirely for this technique. Both tend to treat system prompts as loose suggestions rather than foundational context, which undermines the whole approach.
Sentiment
5 positive (45%) · 2 mixed (18%) · 4 negative (36%)
Trust + time weighted score: +14% · raw score 9%
What redditors said (10 of 11)
Gemini 3 is great, but it seems that right now, only Code offers it, not VS26.
Gemini 3 is better tuned to natural language and collaborative communication. I have already seen better results from this as opposed to concise direction. The fact that it will tell me that it isn't sure rather than coming up with a falicy is huge, the fact that it will ask me for clarification before continuing because something I said may be unclear or could lead to unnecessary interpretation.
Too many people are approaching Gemini 3 as a hammer still, and they keep smacking it's head against a nail and questioning why the nail drives through the head and not the wood.
But the real power of the 2025 models (like Gemini 3 and GPT-5.2) isn't in their "knowledge", it's in their ability to perform **System 2 thinking**
caveat: I haven't tried Gemini 3 for poetic content; so may be good or bad can't vouch
Gemeni is meh.
Gemini 3 is showing improvement, but in my opinion, it’s still not there yet.
However, I'd avoid the GPT-5 family and Gemini 3 entirely for this technique. Both tend to treat system prompts as loose suggestions rather than foundational context, which undermines the whole approach.
Google’s been sleepy. Very sleepy. Gemini 3? Weak. Limp. Low-energy.
So I brought in the best prompt engineer I know: Gemini. Using the map context, I ask Gemini 3 to generate a detailed structured prompt as JSON: camera settings, composition, lighting, quality, everything.
Compare
Gemini 3 vs NotebookLMGemini 3 vs ClaudeGemini 3 vs Axis Atlas
Related
- Gemini · parent · 376 mentions
- Gemini Pro · sibling · 25 mentions
- Gemini 2.5 Pro · sibling · 22 mentions
- ChatGPT · related · 1,159 mentions
- Claude · related · 560 mentions
- GitHub Copilot · related · 223 mentions
- Google Drive · related · 79 mentions
- Perplexity AI · related · 73 mentions