←back to thread

504 points Terretta | 6 comments | | HN request time: 0.839s | source | bottom
Show context
NitpickLawyer ◴[] No.45066063[source]
Tested this yesterday with Cline. It's fast, works well with agentic flows, and produces decent code. No idea why this thread is so negative (also got flagged while I was typing this?) but it's a decent model. I'd say it's at or above gpt5-mini level, which is awesome in my book (I've been maining gpt5-mini for a few weeks now, does the job on a budget).

Things I noted:

- It's fast. I tested it in EU tz, so ymmv

- It does agentic in an interesting way. Instead of editing a file whole or in many places, it does many small passes.

- Had a feature take ~110k tokens (parsing html w/ bs4). Still finished the task. Didn't notice any problems at high context.

- When things didn't work first try, it created a new file to test, did all the mocking / testing there, and then once it worked edited the main module file. Nice. GPT5-mini would often times edit working files, and then get confused and fail the task.

All in all, not bad. At the price point it's at, I could see it as a daily driver. Even agentic stuff w/ opus + gpt5 high as planners and this thing as an implementer. It's fast enough that it might be worth setting it up in parallel and basically replicate pass@x from research.

IMO it's good to have options at every level. Having many providers fight for the market is good, it keeps them on their toes, and brings prices down. GPT5-mini is at 2$/MTok, this is at 1.5$/MTok. This is basically "free", in the great scheme of things. I ndon't get the negativity.

replies(10): >>45066728 #>>45067116 #>>45067311 #>>45067436 #>>45067602 #>>45067936 #>>45068543 #>>45068653 #>>45068788 #>>45074597 #
1. arevno ◴[] No.45068653[source]
> No idea why this thread is so negative

Because politics, as it always has been, is the mind-killer.

HN is incapable of separating the product from the man.

Human nature. It is what it is.

replies(4): >>45070898 #>>45071250 #>>45071390 #>>45071748 #
2. loktarogar ◴[] No.45071250[source]
As far as I can tell, this is running from their servers - which means yes, you need to be able to trust the person who ultimately controls this at a bare minimum. Some trust him, some don't - some have good reasons, some don't.

I can evaluate this as it is, but if I was not trusting of a company, I can't then entrust my data to them, and so I can't evaluate a thing as any more than a toy.

3. apu68653 ◴[] No.45071322[source]
Would you use Winston Churchill social network without qualms?
4. ulfw ◴[] No.45071390[source]
People on here are saying they won't use a Chinese AI even if run locally because "largest geopolitical adversary" but are fine using an actual internal right wing cray cray person's server-run AI
5. mapontosevenths ◴[] No.45071748[source]
> HN is incapable of separating the product from the man.

It sounds unreasonable when phrased that way, but it isn't unreasonable at all for two reasons:

1) The man himself is tied intimately with this company, and he has a deep-seated political ideology. It's deeply rooted enough in him that he's already done things which cost the companies he runs millions upon millions of dollars. His top priority is not to you, the user, or even to his businesses, it is to his political agenda.

2) The man is drug user, who appears not to have been incredibly stable before the drugs. There is a non-zero chance that you will build complicated tooling around this only to have it disappear in a few months after Elon goes on a bender and tweets something bad enough to make even the his supporters hate him. That's a big risk.

replies(1): >>45074940 #
6. throw10920 ◴[] No.45074940[source]
Thank you for providing an actual reasoned argument. I didn't completely agree with the initial premise that you can separate a product from its owner/creator anyway, but you gave a solid argument that's making me reconsider my other beliefs.