Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Claude Code May–August 2026 weekly limits promotion

support.claude.com

Post reply on HN
Print view View original

141–150 of 275 posts

Re: Claude Code May–August 2026 weekly limits promotion

#141

Post by alightsoul »

Barbing wrote: post #122
alightsoul wrote: post #103

Earlier quoted context omitted.

There is no evidence regarding distillation. It is impossible to distill a model in just a month which was the gap between fable and Kimi k3. Anthropic wouldn't even keep up with the load. It is just another example of American exceptionalism.

“In one notable technique, their prompts asked Claude to imagine and articulate the internal reasoning behind a completed response and write it out step by step—effectively generating chain-of-thought training data at scale. We also observed tasks in which Claude was used to generate censorship-safe alternatives to politically sensitive queries like questions about dissidents, party leaders, or authoritarianism, like…

That's what any harness could do. By itself it's not evidence of distillation. Just mentioning Deepseek in there is just fear mongering, again American exceptionalism, which creates the belief only the US has the right to own cutting edge technology. It's a disservice to the entire world, and china happens to be the only challenger to that belief. They surely also asked for tasks to avoid mentioning Trump's involvement in the Epstein files and CP, but don't mention those, due to American exceptionalism. Dario has stated that he is an American exceptionalist. He powers Israeli/US bombings in Gaza, but only cares when China does the same.

This is evidence only if you're an American exceptionalist.

Re: Claude Code May–August 2026 weekly limits promotion

#142

Post by discreteevent »

hombre_fatal wrote: post #58
kbrannigan wrote: post #46

Earlier quoted context omitted.

That make sense. Let me ask you this When the project reaches a level of complexity . Do you simply reach for better models or do you reengineer it or does the project scope stop at the egdge of the model's capabilities. I am asking because in my personal projects after a while they becomes a giant messy ball of wires and i basically trust the model to untangle it for me , by the time it untangles properly, I run int…

With Opus 5 / Fable / gpt-5.6 you can simply ask them to fan out subagents to look for ideal architectural simplifications and rank their findings by impact vs confidence. You can swap out "architectural simplification" with performance opportunities, bugs, correctness, etc. I get the orchestrator agent to then itemize it all into a file where I can keep track of which ones I've implemented. The results are pretty as…

> software engineering is more or less solved

But the example you give is of a terminal for which there are copius examples in open source code. How hard is it really for a pattern matching machine to do that?

If I was doing it I would start by forking an existing repo and I might even say then that "software engineering is more or less solved since the open source revolution"

But I don't work on things like that.

Re: Claude Code May–August 2026 weekly limits promotion

#143

Post by throwaway894345 »

WhiteDawn wrote: post #124
lubujackson wrote: post #55

I think the difference between the Anthropic token maximization approach (vibe code all the things!) and OpenAI's focus on efficiency, terseness and token reduction are going to be the defining features of who wins the long-term race. My money is on the more efficient solution. Even if Anthropic can win some benchmarks by using 3x tokens over 3x time, it is a terrible base to build toward the future. Users are no lon…

Not to sound like a mark, I try to not get attached to any of these providers. I've jumped between copilot, claude, gemini and chatgpt since the start of the year. chatgpt wasn't even worth looking at early this year. Anthropic has the smarter models for sure, and seems to be default in corporate. However, the amount of budget you get with GPT as a user is much better, the harness feels more polished, and the models…

I just need a Sol translator to sit in front of Fable and Opus :/

Re: Claude Code May–August 2026 weekly limits promotion

#144

Post by sethops1 »

jryle70 wrote: post #118
jdoe1337halo wrote: post #7

Anthropic seems cooked right now in terms of compute

Their revenue has increased dramatically. The run rate was 9B in Dec, 19B in Mar, 47B in May. It's now at 65B. https://www.bloomberg.com/news/articles/2026-08-17/anthropic...

Annualized. Which is another word for made-up.

Re: Claude Code May–August 2026 weekly limits promotion

#145

Post by alightsoul »

maxdo wrote: post #137
Barbing wrote: post #122

Earlier quoted context omitted.

“In one notable technique, their prompts asked Claude to imagine and articulate the internal reasoning behind a completed response and write it out step by step—effectively generating chain-of-thought training data at scale. We also observed tasks in which Claude was used to generate censorship-safe alternatives to politically sensitive queries like questions about dissidents, party leaders, or authoritarianism, like…

When they do that they pay billions . Unlike China. That’s the order of law vs bunch of companies in a totalitarian government

They are not paying billions anymore. They just switched to buying and then destroying books. Arguably that's worse than training off of Anna's archive which Chinese companies still do. This comment is another example of American exceptionalism.

Re: Claude Code May–August 2026 weekly limits promotion

#148

Post by vanviegen »

throwup238 wrote: post #117
bicepjai wrote: post #101

Earlier quoted context omitted.

I had the opposite experience. Claude models and its harness feel like they are set to eat tokens for everything, especially if it’s ultracode effort. I have seen it spawn 6 agents and eat my 4-hour quota right in front of me. Codex is on point and follows instructions well even with max effort. I like my analogy of Claude being garrulous and Codex being laconic. If I see more limit hits in same sitting session, I ha…

Just before I canceled my 20x Max sub I had a day where I had ~15% of my weekly I was trying to burn. I set it on ultracode and because I had set the max agents 32 for another project and forgot, the five research agents ended up spinning up a total of 26 subagents and burned through the remainder of my weekly in the span of 20 minutes before I noticed and shut it down. “No, not like THAT!”

How can you burn through your weekly budget in 20 minutes? You'd hit your 5 hour budget way before that, right?

Re: Claude Code May–August 2026 weekly limits promotion

#149

Post by GlickWick »

aic0d3r wrote: post #126

consumer hardware is moving towards local ai and the models you can run now locally are already at opus 4.6 level. within 3 years most pc's sold will be ai computers.

The cost and knowledge required to do that is not viable for the average consumer. Same reason the whole OpenClaw thing imploded.

Re: Claude Code May–August 2026 weekly limits promotion

#150

Post by satvikpendem »

maxdo wrote: post #127

I’m supporting you as a company , but since grok 4.6 and codex , it’s just stupid to go with your pricing model . Cursor + grok + router that uses open ai and a bit of anthropic to verify tests etc it’s a no brainer . Why should I pay you money , and have slower model and less intelligence? Oh yeah and I do not care about limits with cursor ultra at all . Unlike with you .

Cursor really is the best deal right now, the amount of usage you get with Grok is amazing and you can use frontier OpenAI or Anthropic models if you ever need to but most code can be written well by Grok.
Post reply on HN