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

A guide to local coding models

aiforswes.com

Post reply on HN
Print view View original

111–120 of 363 posts

Re: A guide to local coding models

#111

Post by cmrdporcupine »

satvikpendem wrote: post #95
cmrdporcupine wrote: post #77

Earlier quoted context omitted.

I've been a software developer for 25 years, and 30ish years in the industry, and have been programming my whole life. I worked at Google for 10 of those years. I work in C++ and Rust. I know how to write code. I don't pay $100 to "vibe code" and "learn to program" or "avoid learning to program." I pay $100 so I can get my personal (open source) projects done faster and more completely without having to hire people w…

I'm talking about the general trend, not the exceptions. How much of the code do you manually write with the 100 dollar subscription? Vibe coding is a descriptive, not a prescriptive, label.

"How much of the code do you manually write"

I review all of it, but hand write little of it. It's bizarre how I've ended up here, but yep.

That said, I wouldn't / don't trust it with something from scratch, I only trust it to do that because I built -- by hand -- a decent foundation for it to start from.

Re: A guide to local coding models

#112

Post by nineteen999 »

joshribakoff wrote: post #106
simonw wrote: post #4

> I realized I looked at this more from the angle of a hobbiest paying for these coding tools. Someone doing little side projects—not someone in a production setting. I did this because I see a lot of people signing up for $100/mo or $200/mo coding subscriptions for personal projects when they likely don’t need to. Are people really doing that? If that's you, know that you can get a LONG way on the $20/month plans fr…

To me, it doesn’t matter how cheap open AI codex is because that tool just burns up tokens, trying to switch to the wrong version of node using NVM on my machine. It spirals in a loop and never makes progress, for me, no matter how explicitly or verbosely i prompt. On the other hand, Claude has been nothing but productive for me. I’m also confused why you don’t assume people have the intelligence to only upgrade when…

I spent about 45 mins trying to get both Claude and ChatGPT to help get Codex running on my machine (WSL2) and on a Linux NUC, they couldn't help me get it working so I gave up and went back to Claude.

Re: A guide to local coding models

#113

Post by wyldfire »

Workaccount2 wrote: post #12

I'm curious what the mental calculus was that a $5k laptop would competitively benchmark against SOTA models for the next 5 years was. Somewhat comically, the author seems to have made it about 2 days. Out of 1,825. I think the real story is the folly of fixating your eyes on shiny new hardware and searching for justifications. I'm too ashamed to admit how many times I've done that dance... Local models are purely fo…

Is that really the case? This summer there was "Frontier AI performance becomes accessible on consumer hardware within a year" [1] which makes me think it's a mistake to discount the open weights models.

[1] https://epoch.ai/data-insights/consumer-gpu-model-gap

Re: A guide to local coding models

#114

Post by deaux »

kxrm wrote: post #88
mudkipdev wrote: post #66

Earlier quoted context omitted.

Claude's $20 plan should be renamed to "trial". Try Opus and you will reach your limit in 10 minutes. With Sonnet, if you aren't clearing the context very often, you'll hit it within a few hours. I'm sympathetic to developers who are using this as their only AI subscription because while I was working on a challenging bug yesterday I reached the limit before it had even diagnosed the problem and had to switch to anot…

> Try Opus and you will reach your limit in 10 minutes. That hasn't been true with Opus 4.5. I usually hit my limit after an hour of intense sessions.

Daily limit? Weekly limit? Hitting a weekly limit after an hour still doesn't seem very productive.

Re: A guide to local coding models

#115

Post by strangescript »

simonw wrote: post #4

> I realized I looked at this more from the angle of a hobbiest paying for these coding tools. Someone doing little side projects—not someone in a production setting. I did this because I see a lot of people signing up for $100/mo or $200/mo coding subscriptions for personal projects when they likely don’t need to. Are people really doing that? If that's you, know that you can get a LONG way on the $20/month plans fr…

this, provided you don't mind hopping around a lot, 5 20 dollar a month accounts will get you way more tokens typically, also good free models will show up from time to time on openrouter

Re: A guide to local coding models

#116

Post by deaux »

alostpuppy wrote: post #104
simonw wrote: post #50

Earlier quoted context omitted.

With Codex it only happened to me once in my 4.5hr session here: https://simonwillison.net/2025/Dec/15/porting-justhtml/ Claude Code is a whole lot less generous though.

For sure. On one project I kept using codex just to see where the wall was. Took a long time.

It helps that Codex is so much slower than Anthropic models, a 4.5 hours Codex session might as well be a 2 hour Claude Code one. I use both extensively FWIW.

Re: A guide to local coding models

#117

Post by deaux »

InsideOutSanta wrote: post #42
baq wrote: post #21

Earlier quoted context omitted.

bit the bullet this week and paid for a month of claude and a month of chatgpt plus. claude seems to have much lower token limits, both aggregate and rate-limited and GPT-5.2 isn't a bad model at all. $20 for claude is not enough even for a hobby project (after one day!), openai looks like it might be.

I feel like a lot of the criticism the GPT-5.x models receive only applies to specific use cases. I prefer these models over Anthropic's because they are less creative and less likely to take freedoms interpreting my prompts. Sonnet 4.5 is great for vibe coding. You can give it a relatively vague prompt and it will take the initiative to interpret it in a reasonable way. This is good for non-programmers who just want…

A lot of the criticism from GPT-5.x models stems from the fact they're dog slow so you end up paying with your own time.

Re: A guide to local coding models

#118

Post by throwthrowuknow »

mudkipdev wrote: post #66
simonw wrote: post #4

> I realized I looked at this more from the angle of a hobbiest paying for these coding tools. Someone doing little side projects—not someone in a production setting. I did this because I see a lot of people signing up for $100/mo or $200/mo coding subscriptions for personal projects when they likely don’t need to. Are people really doing that? If that's you, know that you can get a LONG way on the $20/month plans fr…

Claude's $20 plan should be renamed to "trial". Try Opus and you will reach your limit in 10 minutes. With Sonnet, if you aren't clearing the context very often, you'll hit it within a few hours. I'm sympathetic to developers who are using this as their only AI subscription because while I was working on a challenging bug yesterday I reached the limit before it had even diagnosed the problem and had to switch to anot…

I half agree, but it should be called “Hobbiest” since that’s what it’s good for. 10 minutes is hyperbolic, I average 1h30m even when using plan mode first and front loading the context with dev diaries, git history, milestone documents and important excerpts from previous conversations. Something tells me your modules might be too big and need refactoring. That said, it’s a pain having to wait hours between sessions and jump when the window opens to make sure I stay on schedule and can get three in a day but that works ok for hobby projects since I can do other things in between. I would agree that if you’re using it for work you absolutely need Max so that should be what’s called the Pro plan but what can you do? They chose the names so now we just need to add disclaimers.

Re: A guide to local coding models

#119

Post by deaux »

haritha-j wrote: post #46
simonw wrote: post #4

> I realized I looked at this more from the angle of a hobbiest paying for these coding tools. Someone doing little side projects—not someone in a production setting. I did this because I see a lot of people signing up for $100/mo or $200/mo coding subscriptions for personal projects when they likely don’t need to. Are people really doing that? If that's you, know that you can get a LONG way on the $20/month plans fr…

I’ve been using vs code copilot pro for a few months and never really had any issue, once you hit the limit for one model, you generally still have a bunch more models to choose from. Unless I was vibe coding massive amounts of code without looking to testing, it’s hard to imagine I will run out of all the available pro models.

Copilot Pro works with a total requests budget rather than per-model limits unless something changed. Could you explain?

Re: A guide to local coding models

#120

Post by ethmarks »

NelsonMinar wrote: post #11

"This particular [80B] model is what I’m using with 128GB of RAM". The author then goes on to breezily suggest you try the 4B model instead of you only have 8GB of RAM. With no discussion of exactly what a hit in quality you'll be taking doing that.

This is like if an article titled "A guide to growing your own food instead of buying produce" explained that the author was using a four-acre plot of farmland but suggested that that reader could also use a potted plant instead. Absolutely baffling.
Post reply on HN