News for hackers. Stuff that matters.
Re:System Card: Claude Mythos Preview [pdf] by ninjagoo on Tuesday April 07, 2026 @05:34PM (#47681644)
> We're not reading the same numbers I think.

We must not be.

That's why I listed out the ones where it is barely competitive from @babelfish's table, which itself is extracted from Pg 186 & 187 of the System Card, which has the comparison with Opus 4.6, GPT 5.4 and Gemini 3.1 Pro.

Sure, it may be better than Opus 4.6 on some of those, but barely achieves a small increase over GPT-5.4 on the ones I called out.

Re:System Card: Claude Mythos Preview [pdf] by nimchimpsky on Tuesday April 07, 2026 @05:56PM (#47681846)
barely competitive ? Mythos column is the first column.

You are the only person with this take on hackernews, everyone else "this is a massive a jump". Fwiwi, the data you list shows the biggest jump I remember for mythos

Re:System Card: Claude Mythos Preview [pdf] by devmor on Tuesday April 07, 2026 @06:48PM (#47682344)
The biggest jump in the numbers they quoted is 6%.

Please look at the columns OTHER than Opus as well.

Re:System Card: Claude Mythos Preview [pdf] by josephg on Tuesday April 07, 2026 @07:03PM (#47682452)
> Combined results (Claude Mythos / Claude Opus 4.6 / GPT-5.4 / Gemini 3.1 Pro)

> Terminal-Bench 2.0: 82.0% / 65.4% / 75.1% / 68.5%

> USAMO: 97.6% / 42.3% / 95.2% / 74.4%

> The biggest jump in the numbers they quoted is 6%.

Just in the numbers you quoted, thats a 16.6% jump in terminal-bench and a 55.3% absolute increase in USAMO over their previous Opus 4.6 model.

Re:System Card: Claude Mythos Preview [pdf] by devmor on Tuesday April 07, 2026 @07:13PM (#47682529)
I don’t know if you’re willingly disregarding everything being said to you or there’s a language barrier here.
Re:System Card: Claude Mythos Preview [pdf] by dang on Friday April 10, 2026 @11:29AM (#47719615)
Can you please stop posting comments with personal swipes in them? You've unfortunately been doing it repeatedly. It's not what this site is for, and destroys what it is for.

If you wouldn't mind reviewing https://news.ycombinator.com/newsguidelines.html and taking the intended spirit of the site more to heart, we'd be grateful.

Re:System Card: Claude Mythos Preview [pdf] by devmor on Friday April 10, 2026 @06:31PM (#47724516)
You're right, I apologize for that. I have been responding with annoyance rather than walking away when I receive replies that appear to be ignoring context.
Re:System Card: Claude Mythos Preview [pdf] by dang on Friday April 10, 2026 @11:16PM (#47726978)
Appreciated! and of course, I know it's not easy - believe me I know...
Re:System Card: Claude Mythos Preview [pdf] by nl on Tuesday April 07, 2026 @09:12PM (#47683502)
It's higher than all other models except vs Gemini 3.1 Pro on MMMLU
Re:System Card: Claude Mythos Preview [pdf] by DroneBetter on Wednesday April 08, 2026 @06:45AM (#47688292)
this just in: HN user forgets how sigmoid functions work
Re:System Card: Claude Mythos Preview [pdf] by dang on Friday April 10, 2026 @11:29AM (#47719627)
Please don't respond to a bad comment by breaking the site guidelines yourself. That only makes things worse.

https://news.ycombinator.com/newsguidelines.html

Re:System Card: Claude Mythos Preview [pdf] by nl on Tuesday April 07, 2026 @09:11PM (#47683495)
> barely competitive

It's higher than all other models except vs Gemini 3.1 Pro on MMMLU

MMMLU is generally thought to be maxed out - as it it might not be possible to score higher than those scores.

> Overall, they estimated that 6.5% of questions in MMLU contained an error, suggesting the maximum attainable score was significantly below 100%[1]

Other models get close on GPQA Diamond, but it wouldn't be surprising to anyone if the max possible on that was around the 95% the top models are scoring.

[1] https://en.wikipedia.org/wiki/MMLU

Re:System Card: Claude Mythos Preview [pdf] by lostmsu on Wednesday April 08, 2026 @08:34AM (#47689312)
You are reading the percentages wrong.

Because 100% is maximum, you should be looking at error rates instead. GPT has 25% on Terminal Bench and the new model has 18%, almost 1.4x reduction.