ishurand4

17 ms·

1.
by ishurand4 1mo ago
Its up!
2.
by ishurand4 2mo ago
How long until Gemma 5 hits?
3.
by ishurand4 2mo ago
3.1 Flash Lite has a better pelican that 3.6?
4.
by ishurand4 3mo ago
The only one I see that thinks it is claude other than claude itself is the GLM series.
5.
by ishurand4 3mo ago
Its a quadratic graph. It starts low but not that capable, gets better and more expensive, and then the time comes in which the capability needed is not the ones of the frontier models and then the price goes down on the companies who host
6.
by ishurand4 3mo ago
A myth, not a model that actually exists
7.
by ishurand4 3mo ago
A system "card" made mostly by the model itself.
8.
by ishurand4 3mo ago
Thats why it is a mythos model
9.
by ishurand4 3mo ago
Well, for me at least, I pay more for input (Up to 1M per prompt) than output (usually max 4k-8k)
10.
by ishurand4 3mo ago
Yeah, it is also known as Claude Mythos 5
11.
by ishurand4 4mo ago
What's the token usage at?
12.
by ishurand4 4mo ago
The numbers they show don't matter. "On multi-round coreference/context recall tests (often cited as MRCR or long-text retrieval benchmarks), Opus 4.7 reportedly dropped from roughly 78.3% down to 32.2% compared to Opus 4.6.&
13.
by ishurand4 4mo ago
And anyway, with quantum, there will be no need for frontier companies as you might be able to even run a 1T param model on a consumer quantum computer.
14.
by ishurand4 4mo ago
They just showed the benchmarks it improved on but it regressed on so much more, such as the MCRR benchmark: "On multi-round coreference/context recall tests (often cited as MRCR or long-text retrieval benchmarks), Opus 4.7 report
15.
by ishurand4 4mo ago
Stack Overflow and Reddit are still getting threads. And as AI gets smarter, the questions will also expand.
16.
by ishurand4 4mo ago
But then they have alot more services since msft bought them, making higher chances on goes down