I guess Gemini 3.7 Flash is no longer at the pareto frontier of speed to intelligence.
Well, there’s still price
Accelerating GPT-5.6 Sol Ultrafast
11–20 of 295 posts
Re: Accelerating GPT-5.6 Sol Ultrafast
#12This is really cool. Someone here commented about similarity between this and hardware advancements for AV encode/decode. I think it's only a matter of time before miniaturization can have a thumbnail sized user-replaceable accessory that contains the LLM built onto the hardware. I admit I don't know how any of that works, but would be amazing to experience. Fully local, fully offline, ultra fast local inference bett…
https://chatjimmy.ai/ Is that. Company behind it just got acquired by AMD
Re: Accelerating GPT-5.6 Sol Ultrafast
#13I've been waiting so long for something amazing to come out of the OpenAI and Cerebras collaboration.
> In our evaluations, GPT-5.6 Sol on Ultrafast mode answered all 2,500 HLE questions in 11 hours and 11 minutes. Claude Fable 5 needed 78 hours and 27 minutes, more than three days of continuous compute, to arrive at the same conclusions. In other words, Ultrafast worked through the frontier of human knowledge in a single working day, achieving comparable accuracy nearly 7× faster.
This is actually insane.
Hopefully the release ultrafast of Terra and Luna too.
Re: Accelerating GPT-5.6 Sol Ultrafast
#14Wow, that's even faster than diffusion LLMs but with the Fable-level quality! Congrats!
Re: Accelerating GPT-5.6 Sol Ultrafast
#15I swear that now frontier AI stuff comes out few times a week.
Re: Accelerating GPT-5.6 Sol Ultrafast
#16I've been waiting so long for something amazing to come out of the OpenAI and Cerebras collaboration. > In our evaluations, GPT-5.6 Sol on Ultrafast mode answered all 2,500 HLE questions in 11 hours and 11 minutes. Claude Fable 5 needed 78 hours and 27 minutes, more than three days of continuous compute, to arrive at the same conclusions. In other words, Ultrafast worked through the frontier of human knowledge in a s…
Feels like the 90's again where single threaded speed is improving fast. ASICs and wafer scale rather than node shrinks, but end result to me the consumer feels the same.
Re: Accelerating GPT-5.6 Sol Ultrafast
#17The corresponding OpenAI post https://openai.com/index/previewing-ultrafast/ There is no pricing info, which could mean it's "if you have to ask..." territory or they are simply gauging interest before deciding
They're expanding access to companies that apply for the program and explain their use cases. So it's very real but limited imo.
The stake in the side of cerebras has always been that the economics are pretty poor.
Who knows if they will subsidizes it to mitigate sticker shock, but it's a safe assumption that it will be scarily expensive. However if you are in a "cost is no obstacle, speed is god" position, it will likely be pure magic.
Re: Accelerating GPT-5.6 Sol Ultrafast
#18This kills the crab.
Compilation time will be a genuine bottleneck for slop coding if this becomes the standard generation rate over the next few years. Go, Zig or even C99 with TCC for dev builds, any language that can get you systems-level performance (or close to it) in a dev environment where you can iterate in ms rather than minutes is going to be immensely more appealing than generating a potential prototype in 10 seconds and waiting 15 minutes for it to compile.
Re: Accelerating GPT-5.6 Sol Ultrafast
#19Earlier quoted context omitted.
They're expanding access to companies that apply for the program and explain their use cases. So it's very real but limited imo.
The stake in the side of cerebras has always been that the economics are pretty poor. Who knows if they will subsidizes it to mitigate sticker shock, but it's a safe assumption that it will be scarily expensive. However if you are in a "cost is no obstacle, speed is god" position, it will likely be pure magic.
Can anyone explain why Cerberus needs to be _fast_ instead of _cheap_?
I don't think I understand why they aren't leveraging the increased speed to do batching to serve more customers at a "normal" tok/s.
Is the limitation, even on cerberus, still that the cache can only serve so many concurrent sessions over time? Is there no scaling advantage? I genuinely do not understand how any of this works.
Re: Accelerating GPT-5.6 Sol Ultrafast
#20Fast mode is already 1.5 times faster and 2x more expensive in the Codex subscription plan. If this thing is 14 times faster, then I can imagine running out of my quota in one session.