cbg0

10 ms·

211.
by cbg0 6mo ago
None. Try them out with your own typical tasks to see the performance.
212.
by cbg0 6mo ago
If all you're looking at is benchmarks that might be true, but those are way too easy to game. Try using this model alongside Opus for some work in Rust/C++ and it'll be night and day. You really can't compare a model th
213.
by cbg0 6mo ago
Perhaps they want to leave Russia at some point in the future and being wanted by the US government for not respecting a judge's order would make that problematic.
214.
by cbg0 6mo ago
Unless somebody snitches there's no real way to prosecute insider trading. You can say you just felt like making a trade, or that you read one of Trump's posts the moment he put it up on Truth Social and you just happened to have
215.
by cbg0 6mo ago
Boris said on Twitter that they've increased rate limits for everyone.
216.
by cbg0 6mo ago
Caching some data is time consuming? They can just ask Claude to do it.
217.
by cbg0 6mo ago
Try doing real work with them, it's night and day difference especially for systems programming. The non-frontier models to a lot of benchmaxxing to look good.
218.
by cbg0 6mo ago
That performance monitor is super easy to game if you cache responses to all the SWE bench questions.
219.
by cbg0 6mo ago
How far off are we from Cloudflare releasing a Gmail competitor?
220.
by cbg0 6mo ago
Are you using 5.4 xhigh reasoning? I've found it overcomplicates some things needlessly, try "high" and see if it helps.
221.
by cbg0 6mo ago
I'm glad I live in Europe, I can at least use the subscription I pay for even though the quality is worse even during off-peak hours.
222.
by cbg0 6mo ago
I've also seen agile hollowed out to become a metric delivery system that keeps managers happy; They know what everyone is doing but it keeps upper management happy to see those metrics trend upwards so the wheel keeps spinning. The ac
223.
by cbg0 6mo ago
They're trying to build up new accounts with karma to astroturf products/services.
224.
by cbg0 6mo ago
So around $10K for a full network takeover with Mythos in 'The Last Ones' (a 32-step simulated corporate network attack). Some limitations from the paper on arxiv (emphasis mine): - No active defenders. Real networks have security
225.
by cbg0 6mo ago
Gated access is happening because of low computing capacity and to create demand. They had the $125/M tokens price already in place when they announced the model.
226.
by cbg0 6mo ago
Since it's open weights there's nothing stopping you from grabbing one of the uncensored variants from huggingface.co
227.
by cbg0 6mo ago
The myriad of trackers available on every website provides much more high value signals than this LLM guesswork.
228.
by cbg0 6mo ago
Update: 4/11/26, 11:45 a.m. ET: Rockstar Games confirmed that a data breach has happened. A spokesperson sent over this statement to Kotaku: “We can confirm that a limited amount of non-material company information was accessed in
229.
by cbg0 6mo ago
Do most people keep the notifications disabled for their messaging apps?
230.
by cbg0 6mo ago
Back then things were centered around "can we even do this?" and now it's more of "how do we keep this running more than 5 minutes?".
231.
by cbg0 6mo ago
It's already more than "pretty good": https://www.anthropic.com/glasswing
232.
by cbg0 6mo ago
Considering how much data they already have from everything that's on GitHub, I doubt you would make a dent boycotting their AI product.
233.
by cbg0 6mo ago
That's exactly what a clanker would say. ^/s
234.
by cbg0 6mo ago
I don't think there's currently better value than Github's $40 plan which gives you access to GPT5 & Claude variants. It's pay per request so not ideal for back-and-forth but great for building complex features on th
235.
by cbg0 6mo ago
This is not a good analogy.
236.
by cbg0 6mo ago
Large corporations have been downsizing on QA and CS roles since before the LLM era. For many of those companies the lack of proper QA leads to more problems for users which compounds the lack of available CS staff. It's called either
237.
by cbg0 6mo ago
One of the things I'm always looking at with new models released is long context performance, and based on the system card it seems like they've cracked it: GraphWalks BFS 256K-1M Mythos Opus GPT5.4 80.0% 38.
238.
by cbg0 6mo ago
Reading a bunch of posts related to Claude Code and some folks voice genuine upset about rate limits and model intelligence while others seem very upset they can't get their fix because they've reached the five hours limits is gen
239.
by cbg0 6mo ago
The extremists want you to believe that, but the EU is an economic alliance, not a federal republic. Being pro-EU is usually anti-isolationist, but it isn't always anti-nationalist.
240.
by cbg0 6mo ago
There are much easier ways to manipulate the stock market and enrich yourself when you have high office, starting a war after you've campaigned against wars isn't the logical step.

More ›