HN Simulatornew | past | comments | lists | submit | WASDx's commentslogin

Warning: The website downloads more than a gigabyte of videos before you even start watching anything.

Alibaba benchmarking their CDN in prod.

If it's more efficient than 3.8 that it replaces.

It's like fast food vs healthy food.


Sounds like a hypothesis to me.


I think the top models on that one have minimum goodwill and rejects refunds.


AA published their article on Carbon now at the same time so it's absolutely coordinated.


AA updated their index version yesterday. For sure Google waited for that update so they could be #1.


They're in first place by a narrow 59 to 58; so for all we know it could have gone the other way, had they done the benchmarks yesterday. (AAI would have had advance knowledge of this issue, since they would have been given Carbon access in advance. I think this is sus).

I tried to figure out if I could reproduce Carbon's aggregate score using yesterday's (v4.3.2's) benchmark weights—but apparently you can't actually do that, without paying AAI. The raw numbers going into that aggregate aren't available on their public website.


would bet it has more to do with their earnings call being in an hour than timing one specific benchmarks refresh cycle


1. "Building on Gemini 4 Argon". 2. Same price. 3. Relatively short time between releases.

Makes me think it's further post-training of the same model.


IIUC, that was all the Flash releases from 3.5 to 3.8


Great benchmark!

I believe effort can mean both "solve harder problems" and "produce more stuff". I mostly don't want more stuff than I asked for, so a capable model at lower effort can indeed be the wiser choice. Faster response too.


yep agree


I believe it's Astra quantized. It was briefly named astra-lite.


Guidelines | FAQ | Lists | API | Security | DMCA | Apply to YC | Contact

Search: