I got tired of touching my mouse every time I Google something, so I built a quick Chrome extension to fix it.
j / s / ↓ = next result
Enter = open link
1-9 = jump straight to result #
Grab it here: https://t.co/nx8lnigxQB
I use both @claudeai and @OpenAI models. I can see that GPT Sol 5.6 medium usage burn rate is higher than Opus 5 xhigh (also considering weekly usage). Opus writes much cleaner code. Sol output is full of loopholes and cannot be considered for serious work.
🚨 I tested Claude Pro ($20)
I used Opus 5 at MED settings (forgot to use max)
I am ABSOLUTELY AMAZED!!!
20 mins is what I was thinking... here is what I actually got:
- 44 minutes of usage
- 58% of 5h limit used (7% weekly!)
- $25 API usage
- 32 Million tokens!!!
Meaning!!!!
You'll get ~$1500 worth of API usage for $20!!!
And a TOTAL of 41 HOURS of WORK!!! (in a month)
ABSOLUTE GOAT!!
I was NOT expecting this!
Codex!! Good bye, I don't think I'll ever come back!
Through the VOICE trial, Terry is using his Neuralink implant to help fine-tune a brain-to-voice interface for himself and others who can’t speak. He trained the algorithm first by miming speech as best he could, then by simply thinking the words and hearing them come out in his own natural voice.
Powered by Grok Voice from @SpaceXAI
We're seeing extraordinary results from @typesafeai. Default mode in 𝚏𝚡 is auto, with a safety reviewer analyzing every command.
That reviewer runs on GPT Luna today. Jev is up to 18x faster (p95) *and* more accurate. It's coming to @vercel AI Gateway and likely new default.
@thekitze Time is precious for everyone. AI slop + human supervision is the key to success. Even most capable models write sloppy code. If not corrected early on, it’s gonna cost us time, money and tokens.
@adamlyttleapps This perfectly demonstrates the difference between OpenAI and Claude models. GPT Sol is like a lazy genius. Even though its code is right, it misses the fine details. Claude models go the extra mile and actually care about them.