yeah it's kind of the point. they're pedagogical tools. they give you the full recipe for building it from scratch or fine-tuning it beyond what it is.
but i'm prepared to be surprised, maybe they'll conjure up something standalone useful this time
That's not the model onlyrealcuzzo is talking about. He is talking about the A9B that was just released five or six minutes ago.
Which is not going to compare well to the 3.8 Flash because it's much smaller; designed to fit quite a large model into quite a small space.
Will be interesting to see how it compares to Muse Glimmer, Qwen 3.8 27B and the older Qwen 3.6 35B (and how its image analysis performance compares to Gemma 4 12B QAT, which is still IMO the most interesting open weights model released all year).
I think their entire pitch about local AI is best demonstrated with their own model that they can optimise for it, and the shape of this model seems optimised for the smaller RTX chips. Indeed, this model appears to be part of their launch picture:
This is something I've been noodling at, splitting a monorepo into a bunch of component projects. I think, unless they have access to something that is wildly outside the scope and scale of current API models, the downside of having a monorepo is the context pollution, and there's no way to escape that.
I'm not fully sold either way yet but I think splitting might actually be critical for moderating the blast radius of agent intervention.
Tomek Korbak, one of the employees, who were fired, was the technical liason to METR for the investigation and states: "I was told verbally I was fired because of the way I communicated with METR".
Who made the decision to invite METR is not public knowledge, as far as I know. I imagine that an important decision like this was made on a much higher level in the organization.
Given the time limits placed on METR and the limits on what time frame they could investigate it seems plausible that they weren't meant to uncover as much as they did. The whole Hugging Face situation seems to have torpedoed OpenAI's hopes of IPOing this year so I'm sure there was a desire from investors, the board, or leadership for heads to roll.
Last time I accidentally said something sarcasticly over-ambitious to an LLM, it shipped this popup callout tooltip feature on a PDP-7 Type 340 vector graphics display emulator that shows you the meaning of the drawing you're pointing at, as well as the address of the instruction that drew it.
reply