Agree :) direct link for those curious:

https://artificialanalysis.ai/models#price-cost

https://artificialanalysis.ai/models?models=gpt-6-luna%2Cglm...

Exactly. The only metric you need to look at is the cost per task and take your pick from the models that match your budget there.

Spark is such a bad model, you are far better off using Astra Medium than using Spark.

Also cache price is crucial, especially for coding tasks (which, really, are where you really start to care the most about unbounded costs)

The analysis might be better on Artifical Analysis' however the way the information is presented by terrydjony is so much better. I'm able to get the exact actionable info I want. Whereas on artificial Analysis I'm analyzing and figuring it out.

It's the equivalent of ai slop writing in a data format.

I have all sorts of local compute, and local models fairly capable the Frontier models still way more capable/faster and local electricity consumption is something else. Good thing it is getting colder around here.

I often pair them up, and I have an Astra or Sol work as a supervisor and reviewer while Qwen 27B FP8 or Qwen 3.8 Flash Next implements things. I mostly do it as an experiment, just to see what kind of level of autonomy I can get, and they are slow to getting a decent reviewed outcome despite Qwen27B running at 100+ tps and 3.5-4K prefill rates and Qwen3.8 Next at 40 tps and 1-2K prefill. I've been also using similar approach more with OMP, not just the straight Pi harness. And OMP seems to be slower because it has more guardrails. OMP has an interesting feature where you can assign a better LLM as an advisor, wehere it just sort of monitors the progress and injects guidance. And it definitely helps, but one has to be careful. It actually turns out to be expensive if the cache reads are expensive. I learned it the hard way. Where on Fireworks' API, the cache rates for GLM 5.3 flash are quite a bit more expensive than for DeepSeek, and a simple runs ended up costing me three bucks in oversight. So a better way is to have a Frontier model running a separate tmux pane and just directing it to Wake up every 10 minutes, take a peek at what's going on, review the milestones give feedback and then sleep. This turns out to be pretty decent cost saving strategy when quote needs to be stretched. Paradoxically, OpenAI tightening up their quota allowance once they released Astra actually pushed me into all these sorts of experiments, and it's actually been interesting. I've been exploring all these smaller flash models, and it's been nice. I do like using local LLMs for chore type tasks that are just mostly information gathering, post-session reviews, stuff like that.

On my 64gb m3 max qwen3.8 27b has been great for planning and then letting qwen3.6 35ba3b actually implement the planned changes.

You might be interested in https://prismml.com/news/bonsai-2-27b

If you're on MacOS, with atleast an M3 chip and 32GB, you should look at the splash engine.

GNHF seems exactly what I've been aiming for to handle overnight tasks.

It's _so_ good, I no longer bother with Sonnet and use it locally for everything.

Consider bumping reasoning down to Medium as a default though, I agree with simonw it over thinks https://simonwillison.net/2026/Aug/16/qwen-38-27b/

I do. For local dev work, I'm mostly using jetbrains' Junie, I can swap between a collection of models from google, openai, anthrophic.

I've had more than a few people tell me "oh, it's so much cheaper to use a $20 claude account" or "i've never hit a limit ever using my openai". Inevitably.. I end up reading/hearing "oh, I need to give it another couple hours to start using it again"... I've never hit that with my approach, even if it's costing me a bit more. Being able to work when I want when I have time has some value.

I also have openai and anthropic direct API billing set up for hosted and client projects that need to call out to an LLM service.

I do, 3-4$ a month of deep seek is enough for my usage

I load OpenRouter up and use models like GLM Flash 5.3, DeepSeek Flash 4.1, Luna, etc. And I often have random niche needs where I need a handful of calls for say, a really good image reader like Gemini Flash 3.8 or whatever. You can do a lot with $25 on openrouter or direct to chinese providers. I am cautious about what data I send overseas, but also like... just because it is in China does not inherently mean it is any less secure than a US provider.

I can't remember the last time any real recourse has mattered for companies getting breached or mishandling my data. Their stock just goes up and the govt just shrugs.

I pay $100 for Codex and it last about a day in the weekly limit - mostly Astra and Sol.

Then i got $20 into DeepSeek and i've been using those $20 for two weeks every day now. Use case is automating computer/browser use - Astra is really good at it, but very expensive, Sol and Luna haven't been that great at it, Deepseek as at about 80% of Astra but lasts forever.

I do, but via OpenRouter. Outside of work my use cases are small and cheaper models do great job at those. I noticed even if I "burn tokens like crazy" I still pay less than any subscription available (a few $ a month).

But I guess if I had an agent vibecoding on it's own, I'd go with subscription instantly.

Or get a 20$ subscription from each major provider and have 40$ spare for openrouter credits. That’s what I’m doing.

What actual usage you want ? Because imho the answer is very different if you want code, or computer use, or vision, or text analysis, or ...

The subscription from frontier labs can handle all that (though each has its best strenght, they all perform well on most things), but it's usually a bit of a waste compared to taking the best one for you.

The problem is that subscriptions with AI models get a bit 'vague' on what you get and how (or when) you might be rate-limited.

I am running it on M3 pro. Works great except for prefill speed which makes it slow for many coding tasks. The newer generation of macs are promising but as a cost-sensitive user, I am also looking into cheaper 32 GB gpus from intel, AMD, nvidia. Eventually, I think these class of models will work well for most coding usecases especially given that I certainly want to have some control of the code generated.