I feel like people use medium models thinking they are saving money, but in the end, it ends up wasting more time and tokens theough the increased rate of issues/bugs.

If the highest tier model makes even one or two fewer mistakes that require human intervantion, that almost makes it a no-brainer choice for production.

▲

spike021 2 days ago | parent | next [-]

I've been doing some research lately on my own (aside from the major benchmarking going on) and I've definitely found cases where medium models do much poorly and cost more than the higher models. But on the other hand, if I can give the medium models certain skills or other resources to complete a task then they perform better in cost than the higher model (which, given the same resources, sometimes gets more confused than the medium does).

At least for my use case of evaluations of a set of features.

▲

XCSme 2 days ago | parent [-]

Yup, in my benchmarks usually medium or high do better than xhigh or max

▲

spider-mario 2 days ago | parent [-]

You’re talking about something else (reasoning levels). The parent comment is referring to e.g. GPT Terra as opposed to Luna, Sol or Astra, Claude Sonnet as opposed to Haiku, Opus or Fable, or Gemini Flash/Hydrogen as opposed to Flash-Lite, Argon or Carbon.

	▲	XCSme 2 days ago \| parent [-]
		Oh, good catch. I do use Sol in production, for cases where I need quite good results, but can't afford to run in production the highest tier model due to pricing (so it's like the highest tier model I can afford for that use-case). To be more specific, this is for a support automation, where the model has to be quite accurate in following rules and generate responses, and Luna models or lower-tier models often make bad mistakes and generate wrong responses, so are not really usable for sending data directly to the user.

▲

selicos 2 days ago | parent | prev | next [-]

Even better at the local LLM stage, where QFN can catch a mistake in prior work and patch it without any subscription costs at all. then I can use a cloud model for polish and finish and move on.

I'm building stuff at home with my work focused on managing infra. If anyone wants to sponsor a SOTA model subscription I will check it out. Until then I'll see how flexible a Strix Halo setup and my homelab can be.

▲

Gareth321 2 days ago | parent | prev [-]

Asking people to guess beforehand how much intelligence they require for a task will hopefully be a relic soon. We’re lacking useful routing techniques which accurately pick the right model for the job. Or a high intelligence model which completes simple tasks fast and cheap. Or something else.

	▲	XCSme 2 days ago \| parent [-]
		Yeah, an xhigh models should know a task is easy and respond quicker.