HackerLangs
トップ新着トレンドコメント過去質問紹介求人

217

26 カルマ登録 2 か月前

投稿

Omp

omp.sh
2 ポイント·投稿者 217·3 日前·3 コメント

It Is Back

2 ポイント·投稿者 217·9 日前·2 コメント

Ask HN: Where the h*ck is fable?

2 ポイント·投稿者 217·9 日前·2 コメント

Is there a "is fable up tracker" or notification hook?

2 ポイント·投稿者 217·10 日前·1 コメント

OpenAI to Stagger Release of GPT 5.6 at Request of U.S. Government

velo.xyz
74 ポイント·投稿者 217·15 日前·26 コメント

コメント

217
·昨日·議論
Consensus itself does NOT matter, omp is objectively the best harness for power users yet it has 0 hn posts about it, zero.

You're fully free to use and try anything and without caring about what others think is right
217
·3 日前·議論
good tech
217
·4 日前·議論
while true, the people who will read this and then think twice about implementing and applying things are exactly the people who already doing too much thinking
217
·4 日前·議論
nothing beside the lack of willing is stopping you from removing its internet access before giving it the file system permissions
217
·4 日前·議論
[dead]
217
·7 日前·議論
slop in its purest non ai form
217
·7 日前·議論
i love seeing things i saw on twitter two years ago at the front page of hn man like what are we doing
217
·11 日前·議論
This is kind of like saying grass is green to be honest
217
·15 日前·議論
im crying bro snuck in india
217
·15 日前·議論
not sure if its fully real, the source is twitter and this weird paywalled website, but if it is, fuck.

https://www.theinformation.com/articles/trump-administration...
217
·15 日前·議論
man.
217
·27 日前·議論
oh wow we really have poki / crazygames at home now
217
·先月·議論
So essentially there are 2 models, Mythos and Fable, they have the same weights but Fable is very safety-nerfed, and only ultra authorized companies have access to mythos with full capabilities

Reported benchmarks:

swe-bench verified mythos 5: 95.5%; fable 5: 95.0%

swe-bench pro mythos 5: 80.3%; fable 5: 80.0%

terminal-bench 2.1 mythos 5: 88.0%; fable 5: 84.3%

gpqa diamond mythos 5: 94.1%

riemannbench mythos 5: 55.0%; mythos preview: 43.0%; opus 4.8: 34.0%

arxivmath mythos 5: 78.5%

critpt mythos 5: 28.6%; gpt-5.5: 27.1%; opus 4.8: 20.9%

graphwalks bfs 1m mythos 5: 79.4%; mythos preview: 74.3%; opus 4.8: 68.1%

humanity’s last exam mythos 5: 59.0% without tools; 64.5% with tools

browsecomp mythos 5: 88.0% single-agent; 93.3% multi-agent

osworld-verified mythos/fable: 85.0%

gdp.pdf fable 5: 29.8% strict pass; mythos 5: 87.6% with tools on mean criteria pass

officeqa pro fable 5: 57.9% on databricks’ eval

legal agent benchmark mythos 5: 16.91% all-pass; 92.0% mean criterion-pass

healthbench mythos 5: 62.7%

healthbench professional mythos 5: 66.0%

multilingual gmmlu / milu / include 93.2%; 92.9%; 90.5%

biomysterybench 83.9% human-solvable; 46.1% human-difficult

organic chemistry mythos 5: 90.1%

labbench2 patent questions mythos 5: 79.8%
217
·先月·議論
Oh my god it's actually here
217
·先月·議論
Also why do they just keep doxing people left and right?

Scott Alexander as the most memorable, and then the backlash after they post the backrooms movie creator's house on twitter recently

Just very shortsighted behavior
217
·先月·議論
That is actually not that bad of an idea huh
217
·先月·議論
Can't wait to see a video from a half sloppy channel about this on my youtube front page in roughly 4 business days
217
·先月·議論
First ever argument being "People do not realise how much of a toll it takes on you if you actually care about the environment"

GUYS

PLEASE

The impact of ai on the enviroment is one of the dumbest psyops in history, how can you claim to know start with that after claiming you know the technology and what it is doing?

There are hundreds of reasons to hate ai but this is just NOT it
217
·先月·議論
somewhat relevant longread:

https://paoloanzn.github.io/2026/04/26/agi-will-always-be-on...
217
·2 か月前·議論
quite a fun game to bot! got this:

Game over You ran out of time!

You correctly sorted 8389560 numbers.