mewse-hn

"The first duty of every Starfleet officer is to the truth, whether it's scientific truth, or historical truth, or personal truth! It is the guiding principle upon which Starfleet is based! If you can't find it within yourself to stand up and tell the truth about what happened, you don't deserve to wear that uniform!"

tried running it against a full anime episode, both vocal isolated and raw. it got confused by english words in the theme song and failed spectacularly three times

ugh, translation is worse than passing a parakeet transcript to a cloud llm. disappointing model.

testing it against parakeet right now for ASR, it's slower, heavier, and has a higher error rate. but i'm doing a pass with the embedded translation to see if that's any good, it might be worth it.

e: also i'm testing it with a Q4 quant that might not be full accuracy

2nding parakeet, had great results with a japanese-tuned variant, beat the pants off whisper-kotoba

To me the most interesting chart on artificial analysis is intelligence index vs cost per task. It's like the third chart down and they graph a pareto line showing where the frontier is and every model on the pareto line has been interesting to me. Luna, Mimo-v2.6-pro, Sol. Hell, Argon.

This MiniMax model is clustered with Haiku 5.5 so (combined with the criticism on here that it was lacklustre while free on opencode go) I don't find it interesting, but every announcement has a chance at pushing that frontier.

I was benchmarking affordable LLMs for japanese->english translation recently and the winner was Kimi K3 - there are a ton of interesting models in the wake of the flagships.

I'm finding the idea of spawning then killing Jellyfin over and over again like Groundhog Day really funny

> I disagree with this.

you disagree with freedom of association?

I am also starting to take issue with "we've developed a new cutting edge model, and nobody can use it" announcements. Most recently with Google's Argon, at least they were using it internally and they'd be slowly rolling it out. This one is from a company I've never heard of, they're not releasing weights or offering API access at this point, it feels like a fairly worthless announcement.

max thinking means benchmaxxing i guess

The gemma 4 models are pretty great

Yeah I've been stuck on 4.8 too, the opus 5 problems seem mostly fixed. It obeys and writes concisely rather than running off, doing its own thing, and emitting a word salad.