Five Hands-On Tests of GPT-Astra, Gemini, and Fable
Peter Yang compared GPT-Astra 6.2, Gemini 4, and Fable 5.2 across five types of scenarios: video, gaming, design, writing, and standing by their views when challenged. Fable 5.2 performed best overall and stood out for its creativity; Astra underperformed in analysis and writing but showed improved design aesthetics, while Carbon excelled at writing but was slower, and its design and 3D capabilities still need improvement.
Body
Here's my new video testing GPT-Astra 6.2 head to head against Gemini 4 and Fable 5.2 across 5 use cases.
I tried all 3 last night for video, game, design, writing, and holding their ground when I push back. Surprisingly, Fable did NOT win on everything.
📌 Watch the full breakdown now: https://t.co/ePailW82Zs
Quick review of GPT Astra and Gemini Carbon:
1. Astra feels lazy next to Fable and Carbon.
It did an incomplete analysis of my YouTube stats and wrote an off-topic essay. OpenAI models "had that dog in them" in the 5.6 days, but don't seem to anymore. On the plus side, its design taste has improved.
2. Carbon really impressed me with its writing.
But it's slow and needs work on design and 3D (it stuffs the UX with too much info). Still, it works harder and feels smarter than Astra.
As for Fable 5.2, well it's still the best overall and kept surprising me with its creativity, as you'll see in the video.
Watch it and let me know what you think in the replies!
AI Trends