Usage-based pricing killing your vibe, here's how to roll your own local AI

BBender about 4 hours ago 25 commentsRead Article on theregister.com

FR version is available. Content is displayed in original English for accuracy.

⚡ Community Insights

Discussion Sentiment

55% Positive

Analyzed from 806 words in the discussion.

Discussion (25 Comments)Read Original on HackerNews

_345•about 3 hours ago

It's a seriously degraded experience from a developer's perspective. Okay you've got one local LLM installed finally after configuring everything perfectly, what happens when you want to run a second instance? Now you've blown past your vram and system ram limits, and you're stuck to just one.

Furthermore, the model they recommend doesn't quite reach ~gpt-5.4-mini level performance- that quality dip means you may as well just pay for something like Kimi K2.6 via openrouter if you want a something ~>= sonnet 4.6 in performance as a backup for when you run out of anthropic/openai usage.

xscott•about 3 hours ago

Your point about caliber/quality is fair, but I have been pretty astonished by some of the newer/better models (Gemma 4 variants, GPT-OSS before that).

However, there's not a lot of memory increase to have multiple sessions in parallel with one model. It's an HTTP server, and other than some caching, basically stateless.

iib•about 2 hours ago

Doesn't llama.cpp (or similar) have to evict the kv cache for this, so that performance is degraded when running multiple sessions? Or how do you load a model in memory and then use it in multiple sessions? I am still learning this stuff

0xbadcafebee•about 3 hours ago

Not sure why you got downvoted. 95% of people should be paying for a subscription. It's far cheaper, far more scalable, and far less hassle.

Local AI only makes sense for a couple of use cases:

  - Privacy
  - Constant churning on tokens
  - Latency
  - Availability

Local AI is "cheaper" when you already have the hardware sitting around, like an old MacBook or gaming GPU, or the API cost (subscriptions will all run out if you churn 24/7) is too high to bare. I'm surprised companies are still selling their old MacBooks to employees, when they could be turning them into Beowulf clusters for cheap AI compute on long-running jobs (the cost is just electricity)

If usage-based pricing is killing your vibe, find a cheaper subscription with higher limits. Here's a list of them compared on price-per-request-limit: https://codeberg.org/mutablecc/calculate-ai-cost/src/branch/...

xscott•about 3 hours ago

I think you're right about the cost/benefit trade-off in general, but I do wonder how much "compaction" Codex and Claude do is to keep context fresh and how much is to save (them) runtime costs.

If you've got a 1M token context, but they constantly summarize it down to something much smaller, is it really 1M tokens of benefit? With a local model, you can use all 256k tokens on your own terms. However, I don't have any benchmarks to know.

ls612•about 1 hour ago

I recently set up a Gemma 4 heretic fine tune on my MacBook to prove that I could more than anything else and it is probably around 4o levels of performance imo. Not fit for any real work. That said the fact that 4o was frontier two years ago and today I can equal it on local hardware and uncensored is pretty impressive.

otabdeveloper4•about 1 hour ago

> 95% of people should be paying for a subscription.

Subscription plans are the "first hit is free" plans. Real pricing once subscriptions are phased out in a year or two is gonna be orders of magnitude more.

2ndorderthought•about 3 hours ago

Why are you running 2 instances anyways? If you want that workflow just rent a few ec2 gpu instances and fire away?

vidarh•about 3 hours ago

If you're going to rent a few ec2 gpu instances you might as well funnel things through openrouter. Not that many of us have workflows where trusting an LLM provider is a problem but sending the data to EC2 is not.

As for why, why would you not? Sitting around waiting for a single assistant is inefficient use of time; I tend to have more like 4-10 instances running in parallel.

2ndorderthought•about 2 hours ago

I absolutely see no reason to send company IP, future plans, and current code base to any other company.

I also do not run 10 agents at the same time. There's no way I could keep up with the volume of work from doing that in any meaningful way

jen20•about 3 hours ago

> Not that many of us have workflows where trusting an LLM provider is a problem but sending the data to EC2 is not.

I'd imagine plenty of people have a problem with trusting fly-by-night inference providers or model owners with opt-out policies [1] [2] about training on your data, who would be more than happy to send data to EC2, or even the same models in Amazon Bedrock.

[1]: https://github.blog/news-insights/company-news/updates-to-gi...

[2]: https://help.openai.com/en/articles/5722486-how-your-data-is...

janice1999•about 3 hours ago

A 24GB Nvidia RTX 3090 TI is ~2000 euro.

2ndorderthought•about 3 hours ago

Which is how many months of Claude or Claude + chatgpt when Claude is down? And do you own anything after using those subscriptions? Can you pick and choose from dozens of models and whatever comes next? Can you play video games with your Claude subscription?

beej71•about 3 hours ago

Believe me when I say that I want to run local models, and I do. But in my testing, 24 GB doesn't get you much brainpower.

2ndorderthought•about 2 hours ago

Have you tried the latest qwen3.6 models?

For most of my questions and 8-9b model works great. Upshot is not having chatgpt/meta sell my data or target me with random thoughts later.

efficax•about 2 hours ago

qwen3.6 does a good job locally except it can take 20-30 minutes to respond to a prompt on a mac studio with 32gb of ram.

smcleod•43 minutes ago

Apple Silicon before the M4 does not have matmul instructions which causes the prompt processing to be very slow. It's quite different on the M5, much like using a nvidia GPU

2ndorderthought•about 2 hours ago

Yea you probably do want to use a GPU for models of that size.

I also wonder what quantization you are using? If you haven't tried other quants I really would

efficax•about 2 hours ago

This is qwen3.6:27b-coding-nvfp4. It's only an M1. If they ever ship an M5 studio with 96GB of ram, that's my next upgrade path for the local llm experiments.

You can get work done with them if you have a harness that can drive outcomes without needing feedback (I've been building a tdd red to green agent harness lately that is very effective if given a good plan upfront). So if you can stand waiting a few days to see results that would only take hours with a model deployed to frontier nvidia hardware, you can get results this way.

datadrivenangel•about 2 hours ago

The time delay is the real issue. Much much slower wall clock time.