Hacker News

New stories

Live mirror
30 storiesupdated just nowView source snapshot
  1. DHH at Rails World – Ruby on Rails no longer needed now because AI can make apps(twitter.com/_architected)
    discuss
  2. AI companies don't bribe reporters. They fund fellowships(drjoshcsimmons.com)
    discuss
  3. Human labour is largely invisible to AI(fohlen.dev)
    discuss
  4. Pixel Joint(pixeljoint.com)
    discuss
  5. The Academy for unusually ambitious young people(theacademysf.com)
    discuss
  6. Show HN: Jev-phone – driving iOS and Android apps with jev(github.com/rajmeet)
    discuss
  7. Adventures in using AI to write papers(andrewpwheeler.com)
    discuss
  8. Modern Motherfucking Website(dreamstation.systems)
    discuss
  9. Local JEV, fine tuned in < 30 mins on Mac Air(shmcsensei.github.io)
    discuss
  10. Elizabeth Holmes scheduled to move to halfway house in August 2027(abcnews.com)
    discuss
  11. Unlike many parents, children are fearless around AI(uc.edu)
    discuss
  12. Polling has become a playground for grifters, liars, and frauds(theargumentmag.com)
    discuss
  13. Make Videos Programmatically with React(github.com/remotion-dev)
    discuss
  14. Disney+ and Hulu raise prices by up to 13 percent after doubling profits(arstechnica.com)
    1comments
  15. The Year AI Came for Us: Teaching Entrepreneurship Will Never Be the Same(steveblank.com)
    discuss
  16. What Australia's Hidden Caves Can Teach Us About Martian Terrain(nautil.us)
    discuss
  17. Show HN: Are My ETFs Overvalued?(etf-copilot.com)
    discuss
  18. Mel: A Real Programmer(melsloop.com)
    discuss
  19. Air – nightly discovery scores for 1,800 US small-caps, with a paper trail(aistockradars.com)
    discuss
  20. Show HN: Learn SQL free with no tracking or ads(sql-easy.com)
    discuss
  21. Parenthood is not Instant Gratification(abhchand.me)
    discuss
  22. Invasions of Canada (1812)(wikipedia.org)
    1comments
  23. Magnetic Thermal Transistor(nature.com)
    discuss
  24. Qualcomm Announces Renewed Agreement with Apple(macrumors.com)
    discuss
  25. Tutoring company tells parents to save their money and 'use AI instead'(afr.com)
    discuss
  26. Cloudflare addressed a cross-tenant data exposure vulnerability in Containers(cloudflare.com)
    discuss
  27. Intent, Not Sophistication: The AI Attacker Is on the Record(simonroses.com)
    discuss
  28. Ask HN: How do you read, sign off, verify, manage the volume of generated slop?
    3comments
  29. America's biggest convenience-store chains overcharge consumers(theguardian.com)
    1comments
  30. Evidence: The Missing Runtime Stream (a.k.a. Not Logs or Telemetry)(gluufederation.medium.com)
    discuss

Best LLM for every budget, updated daily

68 pointsby 1h agobestmodelforyourbudget.terrydjony.com
38 comments
50m agoHN ↗

The coding and math tabs seem to be missing the latest models...

36m agoHN ↗

Math in particular is quite far behind. GPT 5.2 is recommended as the best for highest cost. Really?

48m agoHN ↗

Is there a cheaper model than Gemini 3.8 Flash (High) that maybe/kind-of is on-par with it? For me it works really good but hit the limit in two hours tops... last week was the first time I hit the weekly limit and had to wait 4 days... Claude patches OK, but that is also getting drained really fast these days...

47m agoHN ↗

is this on the $20-ish sub? I hear ultra lasts for a very long time

45m agoHN ↗

Yeah, 20... I was going to try Deepseek later on, just chuck 20 in there and see how good it does, and how far I can go as well...

43m agoHN ↗

yeah I enjoy the speed of Gemini, but I also just have the low tier one (I use the 20x Claude and Codex subs for most of my work). For iteration, Gemini is so much fun, but Opus 5.5 is quite fast, as is sol. If you're on a budget, deepseek does look great

39m agoHN ↗

Yeah, I mean, I think in theory I could push it to the next tier on Gemini, but at the same time I wouldn't mind trying something else, since maybe my workloads are not really that smart and I am wasting a lot of computational power on something a cheaper model with similar capabilities can do.

44m agoHN ↗

Ah OK, yeah, I was actually going to go for that earlier, I'll check it out once I get home, thanks!

48m agoHN ↗

1. In real life, most of us use token packages like OpenCode Go etc.

It would be handy to have a site like this one that takes into account the various deals and attempts to calculate the number of tokens per monthly fee for a chosen model. I realize this makes the task a lot more difficult.

2. It would be handy to have a chart like that for the AI hardware that people own. It helps you decide which model to run (resulting in different levels of intelligence and speed). Also difficult to please everyone (preprocessing vs token generation for example) and to keep updated!

I found https://llm-list.com/ yesterday and when I had a detailed look, I quickly found outdated entries, for example looking at GLM 5.3 flash it listed several providers as "free" that weren't free any longer.

47m agoHN ↗

If you maintain this over time, maybe include other sources than just AA and update the design to look less like zero-shot claude styling (I know that font! I know that color! Lol) it's genuinely useful :)

47m agoHN ↗

This definition of cost is not particularly useful. You want cost per task, not cost per 1m tokens. Artificial Analysis does a good job of this.

25m agoHN ↗

Depends on what level of intelligence you're wanting to use. A vanishingly small number of people can or would want to go to the hardware expense of running something like GLM 5.3 Flash, much less something like K3.

And if you want Astra/Fable/Opus frontier level, then there's no option at all.

But if you don't need that, or you don't need speed... That opens up the discussion. I've been impressed even with how Siri's been doing with the Apple Foundation Models in MacOS/iOS 27 given how small they are.

Edit: I can't even fully spec the M5 Ultra Mac Studio you'd need for GLM5.3 Flash since 512GB isn't available yet, but it's already at $9500 for 256GB RAM.

43m agoHN ↗

all I ever wanted is an updated website where I can see the best models I can run on my different devices locally, I don't get why people are throwing money at these companies

9m agoHN ↗

Guessing that most people don't have machines powerful enough to run good-enough models locally.

42m agoHN ↗

It's missing Opus 5.5 which was released over a day ago (and also is clearly on the pareto frontier).

39m agoHN ↗

Don't know if they updated it since your comment, but for me it's on the graph – and on the Pareto frontier indeed.

21m agoHN ↗

It was expert timing on the commenter's and the developer's part. Opus 5.5 still doesn't show under coding though.

21m agoHN ↗

Opus 5.5 is indeed on the Intelligence tab, but I don't see it on the Coding or Math tabs.

38m agoHN ↗

As the time goes on it only becomes harder to differentiate between model capabilities with just one or two numbers. I would love to see some kind of multi-axis placement of all the models on less objective attributes, like wordiness, willingness to give up, an ability to "think ahead" and pre-solve possible problems in code, for example, that I didn't think of or didn't think of talking about, etc etc etc.

For example I've been really enjoying Deepseek v4.1 Flash, it's very "straightforward" to the point of being almost dumb sometimes, but it's absolutely relentless and would solve almost any problem no matter how inefficient the solution is.

No idea how to measure all that, just average CoT length per task is probably a good approximation for some things, but not others.

35m agoHN ↗

Does anyone actually pay API costs out of their own pocket? It's about 10x cheaper to just get a codex or chat gpt subscription, it's so heavily subsidized compared to the API that I'm sure it would be cheaper to use frontier models on a subscription plan rather than paying API prices for deepseek flash.

32m agoHN ↗

Large enterprises pay per token even through a ChatGPT “membership”, for example.

32m agoHN ↗

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.

29m agoHN ↗

I thought Air was JB's multi-model interface? What is Junie? (I see the buttons, but am very confused by JB's AI offerings in general)

Is it worth the ~10x extra cost over the subscriptions? (This is obviously a leading question). Also, I think you can use OpenAI's subcription login with Air, but not Claude's.

28m agoHN ↗

Why not use a codex or claude subscription? If you use the entire usage allotment on the $200 plan it's about $2,000 in equivalent API costs. Switching providers may be valuable but it's quite literally an order of magnitude cheaper.

3m agoHN ↗

I do. Sharing training data with OpenAI gives me a lot of complementary tokens. I go above that but it's still quite economical and I pick the right model for the task (Luna for most).

25m agoHN ↗

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

24m agoHN ↗

I use local LLMs on my Mac Mini.

Otherwise DeepSeek Flash 4.1 is dirt cheap (other "Flash" models are not that expensive either). I pay (very few dollars) out of my own pocket.

There are many things where having an API Key is necessary.

Maybe I’ve missed the boat though: is there now a method to use an api key to access a subscription?

16m agoHN ↗

You can't use an API key on subscriptions, but I've gotten around it using the `codex exec` command to run requests outside the CLI or GUI if you're already authenticated on that machine. Won't work for all cases, but I've never ran into a limitation in my use case of not having an API key.

29m agoHN ↗

You're better off going directly to artificial analysis, this is a feature-poor/misleading/outdated repackaging

Ex. this type of price estimation is quite naive - some models can require 2-3x the number of tokens to achieve the same level of intelligence. Artificial Analysis' own cost per task is a more fair estimation of cost.

25m agoHN ↗

If you have a 24-64GB mac, consider running Qwen3.8 27B locally at night. It's a bit slower to run locally, but if you're sleeping it's less of a problem.

Depending on your memory, you'll need to use the weaker Q4 versions but they still perform well.

It ranks higher than GPT-5.3 Codex (xhigh) or Claude Opus 4.6 (max) so is great for pairing with https://github.com/kunchenguid/gnhf for nightly experimentation, cleanup, or recommendation lists for in the morning.

23m agoHN ↗

Coding and math graphs are very interesting. Extremely cheap models make it into the upper echelon, delivering 90% of the performance for 1% of the price compared to the #1.

23m agoHN ↗

I find it interesting that with the given metric comparison, for coding at min 50 strength, every frontier model brand is from a distinct vendor: Ling, Qwen, Gemini, Muse, Grok, GPT, and Claude in increasing value.

10m agoHN ↗

Cost per token is an extremely naive way to index cost, renders this chart essentially meaningless.

5m agoHN ↗

I'd really like something that's more oriented around subscription fees.

If I want to spend $100 on LLMs next month, what should I do? Get Claude because Opus 5.5/Fable 5.1 are scoring well? Get Grok because 4.7 is supposedly a good mix of competence and cost? Try out a Chinese model? Don't do a subscription at all like this site is saying?