4 ms·
Opus Fast Mode is 30$/150$/M Input/Output cost. Mythos's pricing (from model card) is 25$/125$ Input/Output cost. Based on this I doubt that Mythos pro is too
by einszwei 4mo ago
Opus Fast Mode is 30$/150$/M Input/Output cost.
Mythos's pricing (from model card) is 25$/125$ Input/Output cost.
Based on this I doubt that Mythos pro is too dangerous to release or provides significantly more value.
- hsuduebc2 4mo agoAs far as my understanding goes. It is not a breakthrough model itself but finetuned model with right tools and skills. Fairly similiar to today's coding agents with difference that they are made for software engineering not cyber security.
- ACCount37 4mo agoMythos is the next point on the scaling curve. It has considerably more parameters than most frontier models of today. Which gives it a lot more oomph per token. Is it a "breakthrough" as in "something novel and unexpected"? No. Is it a "breakthrough" as in "something we know works, but made to work on a greater scale"? Very much so.
- hsuduebc2 4mo agoSo it's just a bigger model? Like for example todays 1T models?
- ACCount37 4mo agoSupposedly 10T scale. Literally the next big thing. A bit like what OpenAI tried with GPT-4.5 - but Anthropic actually made it work with MoE, reasoning, tool use, RLVR, etc. It matters because the "g factor" of today's LLMs is at least in part a function of raw scale. Larger models are just smarter - assuming you can handle the training and inference at this increased scale.
- AnimalMuppet 4mo agoSo, realistically, how much further can this go? How many more orders of magnitude?
- ACCount37 4mo ago100T definitely. 1Q maybe. Beyond that, we need new architectures or new types of inference hardware - maybe both. By some estimates, 100T is about what the human brain has when used to its fullest.
- FergusArgyll 4mo agoYou just disingenuously compared it to fast mode which is expensive not due to model strength or size but because you lose all kind of optimizations