ttul

5 ms·

31.
by ttul 2mo ago
Luna is a very capable model - thanks for pointing that out. Terra is the strange one: not cheap enough or intelligent enough to be on the frontier. But Luna sure is.
32.
by ttul 2mo ago
Fable 5 is just straight up a larger model - I'm guessing at this, but there is plenty of evidence online from people far more plugged in than I am. OpenAI is pursuing a strategy that yields greater operating margins and penetration of
33.
by ttul 2mo ago
My sense is that Fable 5 has “taste”. But Sol gets to work and gets shit done. I reserve Fable for when things need a refresh or if I want a flawless front end. Sol does the majority of actual work. I max out two of each at the Max/Pro
34.
by ttul 2mo ago
The DeepSWE benchmark they report (59.3%) overlaps with the confidence interval of 5.6-Sol Medium (61% +/- 2%), but likely at 1/18th the cost (they did not report the DeepSWE benchmark cost, but v4-flash had this cost ratio agains
35.
by ttul 2mo ago
A good share of humanity would have also gotten this question wrong!
36.
by ttul 2mo ago
If you ask Sol or Claude how much time it will take to implement a plan they just came up with, they usually advise a timeframe in the weeks or months - assuming, I suppose, that human programmers will be building it. And then you ask the m
37.
by ttul 2mo ago
Poor BrandonM... He has not been invited to cocktail parties at all since then.
38.
by ttul 2mo ago
Hot water for the whole neighbourhood!
39.
by ttul 2mo ago
Indeed. You need 45 to 60 liters per second of cooling water flowing over a Cerebras wafer every minute to keep it under 90C. And that’s assuming the water leaves at 90C… More realistically, you need much more cooling water.
40.
by ttul 2mo ago
I’ll get that 250kW home power service dropped in next week!
41.
by ttul 2mo ago
(Otherwise it would have slowly transformed into CO2 over the eons without our help)
42.
by ttul 2mo ago
I think there is literally no oxygen down there with the hydrogen. It’s the same with hydraulic fracturing of hydrocarbons. Plenty of methane COULD go boom, but there is no oxygen deep underground.
43.
by ttul 2mo ago
I am waiting with bated breath to read, “load-bearing” somewhere… The latest models are very capable, but sometimes they seem to get so deep in the details that they lose the overall plot. What the hell is the point of this page? Can you pu
44.
by ttul 2mo ago
What is the moment tensor in this context?
45.
by ttul 2mo ago
Try `1 + 0.16 cos(4 t) + 0.08 cos(12 t + pi/6) + 0.04 cos(36 t + pi/3) + 0.02 cos(108 t + pi/2)`
46.
by ttul 2mo ago
With this level of intelligence offered at this level of speed, new real-time applications become possible, such as providing expert advice during a phone call or court hearing. Current SOTA models are too slow in many cases to provide the
47.
by ttul 2mo ago
Always the bringer of facts to rain on my factless fun. Thank you.
48.
by ttul 2mo ago
I’m in a very small C-suite and AI psychosis definitely takes over from my normal level of dysfunction at times. Fortunately, my reports are good at telling me when I am full of shit.
49.
by ttul 2mo ago
2028, everyone. Hang in there.
50.
by ttul 2mo ago
These supernova-generating black holes are atomic scale (Schwartzchild radius of 15pm to 15nm), yet they carry the momentum of a large asteroid. If one of these hit the earth, they would cause a life-ending apocalypse like the event that en
51.
by ttul 2mo ago
We are seeing an exponential increase in the share of site traffic to our B2B company originating from ChatGPT and other models. Google Search is cooked.
52.
by ttul 2mo ago
Google TPUs are built around a 128×128 systolic array of multiply-accumulate (MAC) units. Trainium 1, Trainium 2, and Inferentia 2 also feature a 128x128 systolic array. You learn something every day. Today, it was the term "systolic a
53.
by ttul 3mo ago
It’s relatively easy to get access to the frontier labs’ security programs. This was not always the case. But in the last week, my team got approved for both Anthropic and OpenAI’s programs. They are trying. The labs know that if they don’t
54.
by ttul 3mo ago
Now who wants to say ChatGPT 5.6 Sol is just another token predictor...?
55.
by ttul 3mo ago
That's not the point. Often in medicine, the downside of using one thing is that you're not using another thing that would be safer and more effective. When people chase results from unproven therapies, often it means they're
56.
by ttul 3mo ago
What value is there to an informal conclusion when the topic is as important as intelligence? The downsides to creatine if taken without good scientific backing should not be disregarded, yet that’s exactly what many ordinary people are now
57.
by ttul 3mo ago
If the weights are open, censorship can be easily trained out of the Chinese models. But if you’re sending your tokens to China, all bets are off!
58.
by ttul 3mo ago
Oh yes, I know it’s a harness feature.
59.
by ttul 3mo ago
Fable seems to be a larger model. It costs more to run and does not seem superior for _typical_ software engineering work. But for work requiring raw intelligence, perhaps its size is an advantage. On the DeepSWE 1.1 benchmark (IMHO current
60.
by ttul 3mo ago
I think ultra mode needs to be more clearly documented (or perhaps cautioned against!). Most devs - myself included - who saw “ultra” mode figured that it’s just a magic bullet that makes the model work harder and achieve better results. Bu

More ›