Live data from Hacker News

Kimi K3, and what we can still learn from the pelican benchmark

simonwillison.net

241–245 of 245 posts

Re: Kimi K3, and what we can still learn from the pelican benchmark

#241

Like Simon concludes the article, the main use of this isn't to say which model is "better", but to try and poke at the model to sort out things like quality vs cost vs speed. So I put together a quick comparison of the last couple iterations of Opus, Fable and now Kimi. Kimi is cheapest by 5x but also slowest by 2x https://9gpyw4uxr2.evvl.io/

Personally I'd consider the three middle ones to be failing, in the typical "Gemini/Google" fashion in that the model is doing more than what the prompt asks. The prompt asks for SVG, yet the model is providing more. Edit: Actually, looking at the K2.6 response, that's borderline failing too, it's using HTML+CSS+SVG, not just SVG, again failing to follow the prompt properly. By the way, that website seems like a blac…

It's a free site, so I was trying to limit both the privacy and risk exposure.

Making the content auto expire after a short period of time greatly decreases the attractiveness of the site to lots of SEO spammers and other types of abuse, and if someone were to get something malicious or vile posted it will clean up after itself without me having to wade into things.

Re: Kimi K3, and what we can still learn from the pelican benchmark

#243

Earlier quoted context omitted.

You did attack him. Calling his content spam is as personal as me calling you out for it in harsh language. Implying that he gets special treatment because of ideology instead of the value of his work is incredibly personal. It does feel right to respond in kind to an anonymous troll attacking someone working in public for free. If you had opened the conversation with a well phrased criticism or question (“Simon, are…

> You aren’t a moderator Are you? Really, it's "impressive" that you got to write such a massive wall of nothing, with so much hate and aggressiveness to defend a spammer with zero arguments other than you thinking him posting dozens of links daily isn't spamming. You spent a lot of time stalking and harassing someone because you got annoyed by them calling out an influencer you like. This is the real shitposting. A…

You're right. My point has been made.

To the non-trolls: don't be discouraged by people who have nothing but negativity to add. Do something great. Build people up. Be proud of your work. Share it. There will always be a person who has a problem with what you are doing. Ignore that asshole.

Back to you brazukadev, I assume you'll want to get the last word in.

Re: Kimi K3, and what we can still learn from the pelican benchmark

#244

Earlier quoted context omitted.

> You aren’t a moderator Are you? Really, it's "impressive" that you got to write such a massive wall of nothing, with so much hate and aggressiveness to defend a spammer with zero arguments other than you thinking him posting dozens of links daily isn't spamming. You spent a lot of time stalking and harassing someone because you got annoyed by them calling out an influencer you like. This is the real shitposting. A…

You're right. My point has been made. To the non-trolls: don't be discouraged by people who have nothing but negativity to add. Do something great. Build people up. Be proud of your work. Share it. There will always be a person who has a problem with what you are doing. Ignore that asshole. Back to you brazukadev, I assume you'll want to get the last word in.

You are a very funny negative troll with too much free time.

Pretending to be motivating small creators by helping paid influencers hack their way to the top of every thread.

Re: Kimi K3, and what we can still learn from the pelican benchmark

#245

we can learn nothing from it apart from the large troll community that is HN that wants to do the same boring spiel every time a new model drops

don't blame the community for the work of one hustler and a permissive (just in this case) moderation.

now u mention it, yeah, you’re right
Post reply on HN