guy
499 posts
@TradersConf couldn’t you just scale up if you’re consistently profitable? Lots of grifters in this space but for this specific example it’s a mindset problem rather than a “bad” comparison no?
English
Trading Twitter convinced everyone $30k/month is normal. If you're consistently making $3-5k/month from trading you're doing better than 99% of people who try. Stop comparing yourself to people who are probably lying.
English
I still don't understand what the purple one is used for 😭
can someone explain?
English
look at this fuckwit
Ben Smoke@bencsmoke
they kicked him repeatedly in the head when he was already on the ground and incapacitated…
English
“Guys, don’t kick someone who stabbed two people!”
How are these people allowed to vote?
John Wight@JohnWight1
This is outrageous. The guy is clearly incapacitated on the ground, having rightfully and thankfully been tased, yet two police officers are repeatedly kicking him in the head. Who do they think they are: members of the IDF?
English
my biggest issue with Codex is it somehow leaks implementation or chat details in the frontend?
For example, I told it I will give someone on my team (Ella) access to a tool it’s building so she can manage the data
When prompted to make a landing page, it had a line about how the tool is “made for Ella, , , etc.”
Another example is when something is loading, it just adds a message about exactly what is happening (the OpenAI call is doing XXX, the docker container is spinning up, etc) in the frontend under the loader
Anyone else experiencing this?
English
Traders read charts. But can you read luck?
Pick a number between 1 and 50 👇
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
One number hides a surprise.
Drop it in the comments. 24 hours on the clock. ⏳
English
@yacineMTB @VictorTaelin that’s how I work with all the semi or full agentic tools nowadays - I do the thinking and I delegate the grunt work to the AI
English
@VictorTaelin I find 5.5 more useful for the things I want it to do; which is low intelligence low thinking work that it needs to be diligent and exhaustive about (a lot of my work is not actually high intelligence)
English
DeepSeek is the best OSS model on LamBench . . .
That said, it is still not SOTA. I think Chinese labs are doing poorly because this is a new bench that they couldn't max for. These results align well with how smart they feel to me.
I'm rooting for them though 😕
I just wanna be free from Anthropic...
Also, Opus 4.6 > 4.7 and GPT 5.4 > 5.5 align with my experience. This whole bench captures my feelings extraordinarily well, and I did nothing other than write a bunch of problems and score the models...
Problems available on: VictorTaelin / LamBench
English
@markbuildsbrand I think that the “hyper leap” will be less about the LLM itself and more about how vendors make the LLMs easier to harness in creative ways - we’re going to see a shift from chatbots I think
English
the benchmarks show GPT-5.5 blows Opus out of the water...
i don't see it.
just a tad bit faster and 2x as expensive.
Opus 4.7 was marginal gain over 4.6 as well.
relatively underwhelming updates recently imo. I want another "holy shit" update from anthropic or OpenAI
OpenAI@OpenAI
Introducing GPT-5.5 A new class of intelligence for real work and powering agents, built to understand complex goals, use tools, check its work, and carry more tasks through to completion. It marks a new way of getting computer work done. Now available in ChatGPT and Codex.
English
months later, I think we all see what the actual bubble is
guy@guytrvs
seeing so many retarded headlines about the ai bubble if you have any ounce of critical thinking you realise how false this is
English
have to say @oliverbrocato has to be one of the goats he built a massive ecom brand, sold it, then built a solution to his biggest problem
huge respect
English