MiniMax open sources M2
This model has been shaking the benchmarks last week, now that it’s open we see that it’s 230B-A10B and dueling (arguably beating) Sonnet 4.5 at 8% of the cost
github.com/MiniMax-AI/M...
MiniMax open sources M2
View original thread
34
3
why build a model? vertical integration
seems like US models actually do work fine, they’re just too expensive
also, a little bit of an admission that Chinese models are a little wonky
seems like US models actually do work fine, they’re just too expensive
also, a little bit of an admission that Chinese models are a little wonky
4
how they’re thinking about model architecture
2
this is another model in fp8 format, definitely a trend toward lower precision
gpt-oss was in fp4, but several others have been in fp8. i’m a bit surprised that attention is in fp8, but the model does seem to work well
1M context (if you can pay for the KV cache)
gpt-oss was in fp4, but several others have been in fp8. i’m a bit surprised that attention is in fp8, but the model does seem to work well
1M context (if you can pay for the KV cache)
3
gents, i think we’ve reached it
9
