zlacker

[parent] [thread] 3 comments
1. arnaud+(OP)[view] [source] 2025-05-06 16:10:25
Be careful, this model is worse than 03-25 in 10 of the 12 benchmarks (!)

I bet they kept training on coding, made everything worse on the way, and tried to hide it under the rug because of the sunk costs.

replies(2): >>jstumm+mc >>ramble+4N1
◧
2. jstumm+mc[view] [source] 2025-05-06 17:23:30
>>arnaud+(OP)
It seems that trying to build llms is the definition of accepting sunk cost.
◧
3. ramble+4N1[view] [source] 2025-05-07 09:20:12
>>arnaud+(OP)
where do you see that?
replies(1): >>arnaud+TU1
◧◩
4. arnaud+TU1[view] [source] [discussion] 2025-05-07 10:46:08
>>ramble+4N1
New model homepage : https://deepmind.google/technologies/gemini/

Old model card : https://storage.googleapis.com/model-cards/documents/gemini-...

They intentionally buried that information

[go to top]