The mysterious Hy3 LLM is topping OpenRouter Model Rankings by a large margin

andai

So basically, Hy3 is the cheapest decent model on OpenRouter, unless you use DeepSeek as the provider for DeepSeek V4 Flash, in which case DeepSeek's insane caching wins out. (And Hy3 is close-ish on the benchmarks.)

Aurornis

> Two new models are now beating LLM darling Claude in terms of token usage and by more than 50%? Time for a reminder that OpenRouter leaderboards only show tokens sent through OpenRouter, which most Anthropic API users don’t use.

0xbadcafebee

> it makes sense that a cheaper model would prevail, but only if it offered similar quality You're trying to think logically, which has no place in an AI discussion. :) People just jump to whatever the latest model is. Plenty of people also prefer price to "quality" (which is very subjective). It's new, it's cheap, so people use it. It's likely people will stop using it when something else is cheaper and/or newer.

vessenes

Since there’s only one inference provider it could be a recycling/ad experiment. The similar usage between trial and paid periods would be explained by this as well.

simonw

First model I've tried that gave me back HTML with a "Change Pelican Color" button: https://static.simonwillison.net/static/2026/hy3-preview-pel... (Transcript: https://gist.github.com/simonw/c2a0d8ecd3056a2681319eae8fc3f... )

simonw

OpenRouter rankings frustrate me, because they show the total number of tokens but they provide no indication of how many unique users a model has. Which means if a surprise model tops the leaderboard one week we can never be sure if it was because a single whale user pushing billions of tokens a day switched to it, or if it represents a genuine community trend towards that model.

bandrami

For the life of me I will never understand the thought process that leads you to say "we don't really know who developed this LLM but I'm going to feed all of my business's data to it"

freakynit

This was originally a 400+B param model which was later reduced to 295B considering it as the "optimal zone". https://www.mdshare.online/s/uend0pj3og_A_rgcxzINf

zone411

I’ve tested this model on four of my benchmarks: https://github.com/lechmazur/buyout_game 10th out 36. https://github.com/lechmazur/pact/ 14th out 25. https://github.com/lechmazur/nyt-connections/ 60th out 81. https://github.com/lechmazur/debate 16th out of 29.

thot_experiment

Tried this extensively in OpenCode, never used it once since Gemma 4 came out, got into thought loops and did stupid edits I didn't ask for more often than the local 31b model. One of the worst "frontier" models I've ever tried.