Argon - what a pirate says when they’ve made all of the captive nobility walk the plank.
2026-10-07 14:39:23
0
Shladee :
Gemini so bad at the moment and is really frustrating to use compared to the other models
2026-10-07 01:05:35
4
kevinsmak :
I don’t think I need anything better than opus 5.5 right now. Then a new model comes out and usually changes my mind..
2026-10-07 12:33:11
1
Dan.about.ai :
Calling the model Argonne because it supposedly means “doing nothing” is a pretty awkward launch coincidence.
2026-10-07 13:28:22
0
tzar0n :
doesnt Google have a quantum computer?if so they will trouple everyone.
2026-10-07 03:28:09
1
sabotaged209 :
grok suddenly became relevant again, so anything is possible
2026-10-07 06:59:37
3
Finn 🍉 :
I don't want nor need a smarter model. I need the models to become cheaper and smaller in parameter size. My Max plan now hits weekly limit after 3-4 days, compared to 5-6 days a month ago.
2026-10-07 00:00:07
15
Jinxycat :
If you combine the Jevons Paradox with Flywheel economics, the answer is a clear no. The gap between the frontier and later adopters or movers compounds. The gap dictates who captures the demand created by Jevons.
2026-10-07 07:34:34
1
stefprice492 :
Its getting impossible to keep up with all these model releases.
2026-10-07 05:30:12
2
Juan Daniel Reich :
I'm using Opus 5.5 on High and it makes mistakes, then my specialized QA agent catches some of the mistakes that... Well, he made himself and then QA introduces new errors when it fixes the first ones, so...
The whole "they're crazy smart" I don't find them that smart.
2026-10-07 00:11:22
3
SilverSurfer :
wolfram
2026-10-07 03:14:46
1
kiki • a special sparkle ✌🏼✨ :
Fable 5.5 ?! RIP my token budget
2026-10-07 05:06:34
1
peter kerr :
then what does Argonaut mean?
2026-10-07 01:57:49
0
cjjames5021 :
Not holding my breath
2026-10-07 01:44:40
1
Nextlesson.com :
They are smart keeping things super cheap at scale. They should find some way to make a Muse that can work with one of their flash models.
2026-10-07 04:01:33
0
rageagainst :
All these companies chasing benchmarks when they should be working on the software layer on top. Most people are using AI for fairly simple tasks where since Opus 4.5 we have only seen marginal improvements. We need improvements to usability, better plugins and more efficient workflows. Good example of this has been the docs and slides feature in Claude where you can now Cowork on slides with the model in real time as they are being made
2026-10-07 01:14:45
1
chrisgrey888 :
I'm curious how it will compare to Grok 4.7. From what I understand, none of the Groks have been worth using until 4.7. And once they get Cursor's RL enhancements built into Grok, I'm hoping it will be a good replacement for Composer 2.5. But the rumor is Cursor will get Composer 3 before we get a Grok model as the default.
2026-10-07 00:56:58
1
Grant Steven :
When you use it for basic search, it will be just fine
2026-10-07 02:58:28
0
eternal perspective :
artilect wars incoming- God help us 💀
2026-10-06 23:59:03
0
r9999 :
I use both Claude Fable 5.1 and Google Gemini 3.8 Pro often. most people think Claude is better and everything but it's not always the case! just the other day I went three rounds with Claude trying to get an answer for something but no luck and Gemini did it in one shot!
2026-10-07 02:02:20
0
taptapdrver🇭🇹🇯🇲 :
Google have alot of products around Gemini
2026-10-07 00:33:10
0
TheActualTruth :
It’s not solving long dead equations
2026-10-07 00:32:29
0
TheActualTruth :
Catch up to what?
2026-10-07 00:32:03
0
you :
coming when it's been 2 weeks
2026-10-07 00:08:15
0
PapayaPhil :
Nate what's your take on AI reverse engineering 99% of any software. this past week the gaming modding community has done a lot of game merging
2026-10-07 00:26:03
0
To see more videos from user @nate.b.jones, please go to the Tikwm
homepage.