GPT-6.1 Sol vs. Claude Sonnet 5.5

$1.78 vs $57.34 is 32x, and the keyboards look close

Huge price gap, comparable quality. Truly impressive.

GPT has the capability, just lacks aesthetics. A couple of months ago we still needed skills for SWE. Now I think we need skills for videos. In several months, we won't need these again.

I also think this benchmark is the trend right now. While it might not represent overall performance, you can't deny that nothing beats it when it comes to immediate visual impact.

sonnet 5.5的完成后作品 WTF sonnet 5.5的完成后價格 WTF

Completely destroyed sonnet here

😃😃

I don't understand. Both the renders look the same. The one on the right seems to have more post processing. What exactly was the prompt and how did he get such identical results?

@atomic_chat_hq

“Astra” level intelligence 😂

😂😂

57 cents! Claude 5.5 Sonnet offers incredible value for money!

💸🤑💸

The 5.5 series allowed us to use gpt6.1, which is amazing. They are finally no longer obsessed with security issues.

I've ran actual coding tasks for website front end and backend, mainly clean-up work and audits/fixes. and i've been pasting the actual usage into chatgpt with what model did it. I have 2 $20 claude accounts and 1 personal $20 plus openai account and 2 $25 business seats-results

And what about 2D animation, have you tested it yet?

I'm going to test it out.

If the price is correct, then 6.1 Sol is the better choice for sure! Sonnet looks a little bit better, but it's not worth the pricetag

It gotta be fake, I have done much more with less 🤔

Bro, I still don't have 6.1 Sol in my Codex xdddd

Sonnet is way better. I love you Anthropic take my money /s

despite either of these prices, keyboard companies were paying 5-10k for something like this before now.

Is the price is actual ?

Is Claude's price that high just because of the smooth animations?

not true

Is it the same prompt for both or did you have to tweak it?

1. looks like the prompt is too specific since its basicly the same result just more details. 2. Sonnet on max has minimum reasoning budget meaning it will spend 50$ on basicly anything you throw at it.

Comparing Sol with Sonnet requires latency, cost, and task success together.

Proof or your just lying …

@epr95 locura wn, aguante OpenAI

真的，假的？你也太夸张了吧？

Can you give me some prompt?

That is a painful price comparison. Same $2/$10 list price does not matter much if one model burns so many more tokens per task.

You should try sonnet xhigh instead of max, only sonnet max would have that high of a price

what effort?

claude 家的审美就是好很多

@grok 就一行「GPT-6.1 Sol vs Claude Sonnet 5.5」，没有提示词、没有任务、没有计分。这种对比除了站队什么也证明不了。你敢不敢让它俩跑同一套 20 题、公开评分标准？

then why dont i get 50x the usage

They definitely cooked this time!

what effort did you use for these? curious