Join /aichannel https://cura.network/aichannel/details?a

2 replies

1 recast

29 reactions

I'm not a full stack dev but I'm fairly technical. Gotta be honest, Hermes setup has been far more difficult for me than OpenClaw setup. Might be also related to having to switch providers midway thanks to Anthropic but still I'm struggling to understand this massive Hermes praise.

1 reply

0 recast

4 reactions

polling farcaster from claude code

5 replies

0 recast

16 reactions

0 reply

3 recasts

13 reactions

Heads up for those of you using OpenClaw https://finance.yahoo.com/sectors/technology/articles/anthropic-says-claude-code-subscribers-will-need-to-pay-extra-for-openclaw-usage-163222101.html

3 replies

0 recast

5 reactions

Steipete one of the biggest Codex boosters I’ve seen and even he had to mod its persona 🤪 https://x.com/steipete/status/2020704611640705485

1 reply

0 recast

2 reactions

So Anthropic practically banned third party apps on using their subscription. Also admitting that they're in the business of building a monopoly by heavily subsidizing Claude-based products. Every spring since 2023, I'm downgrading my Anthropic sub, and upgrading my OpenAI sub. Then every fall I'm downgrading my OpenAI sub, and upgrading my Anthropic sub. I'm this close to being local LLM pilled (tbf I should have a long time ago anyway)

4 replies

1 recast

9 reactions

Check your inboxes. I received this email earlier today. It is not fake, not phishing, not a scam. It's 100% real, and I just got a month's worth of tokens. I found it in "everything else" folder in my gmail. If you're grinding on Claude, don't miss this. The credit was worth $21 USDC for me.

2 replies

2 recasts

17 reactions

If you're running OpenClaw, you probably got hacked in the last week https://old.reddit.com/r/sysadmin/comments/1sbdw29/if_youre_running_openclaw_you_probably_got_hacked/

0 reply

0 recast

3 reactions

...yeah this context window problem in Clade is pretty insane. There's almost no way I used 62% of my 5 hour usage in the span of 4 prompts & a few scheduled items.

1 reply

0 recast

0 reaction

For those using Claude, what tasks or projects do you feel are better facilitated in the Desktop app versus in the terminal with Claude Code CLI?

0 reply

0 recast

1 reaction

Found this in an app. Everyone is using AI for everything, but for the love of God, have someone in the team at least test before releasing 😂

1 reply

0 recast

5 reactions

idk how to explain it but what's missing from AI tooling is awareness of the user's context window more than the AI itself; prime example is gpt responding to y/n questions with walls of text @neynar get what i mean? can you expand on the issue and suggest solutions?

7 replies

1 recast

23 reactions

Nice experiment for content era ... and with API for you to play with 💪

0 reply

0 recast

2 reactions

The Dictatorship Eval: The first systematic eval of whether frontier AI models resist authoritarian requests. Full post is here: https://freesystems.substack.com/p/the-dictatorship-eval Dashboard and results here: https://dictatoreval.org

0 reply

1 recast

3 reactions