Introducing SubQ - a major breakthrough in LLM intelligence.
@alex_whedon26.177 Follower
13.499.800 Aufrufe
70 Kommentare
Kommentare aus dem Originalbeitrag auf X (Twitter)

@alex_whedon
SubQ is available for early access today, alongside our coding agent, SubQ Code Get access today ↓
1.798 Likes

@alex_whedon
We were a little slow on this, but we just got a technical blog post up with more details. Please take a look! We have a model card coming next week, and we are happy to take requests for any specific details there. I am happy to answer any questions here!
680 Likes

@MartinShkreli
congrats!! why only 150tok/sec if 52x faster, etc.?
361 Likes

@alex_whedon
52x is for prefill speed. Decoding speedups are coming too! This is the first research announcement, with many to follow.
426 Likes

@chrisbward
@MartinShkreli absolute grift, didn't deliver
1 Like

@alex_whedon
@MartinShkreli Model card shared!

@vincexplorer
Any papers? Seems too good to be true
318 Likes

@alex_whedon
The model card is coming next week! We are releasing a technical blog post with more details later today.
414 Likes

@vincexplorer
Looking forward to reading it
40 Likes

@shiri_shh
i still can't believe this.. 12M context window with 98% accuracy 💀 @alex_whedon @subquadratic
301 Likes

@alex_whedon
@subquadratic And at a fraction of the cost...
218 Likes

@NanouuSymeon
If this works as described, it basically changes how we build with LLMs. A lot of current pipelines exist just to work around context limits.
142 Likes

@alex_whedon
And now that constraint goes away, what people choose to build is going to be very different.
98 Likes

@LinusEkenstam
I remember when Gemini first made the claim of 1M in lab situations. We've come quite a bit from there. I'm extremely intrigued to see the real-world implications of this in bigger and more complex AI powered software. There is the zero shot environment, but there is also the qualitative repetitive work that is prone to drift and degradation over time. I really hope you will share case-studies, since I think this will be one of the most impressive ways to highlight the capabilities. Amazing work Alex and team. Kudos.
116 Likes

@alex_whedon
This is the framing we'd love to test against. If you've got a workload that historically drifts, would genuinely want you to throw it at SubQ and see what happens.
72 Likes

@Den_zel21
If SubQ only attends to a small subset of tokens, how does it guarantee recall of critical long range dependencies in worst case inputs (e.g adversarial prompts or tasks where the important tokens aren’t locally obvious)
108 Likes

@alex_whedon
We dynamically select the token relationships that matter and train it against retrieval problems with distractor docs, etc.
108 Likes

@Den_zel21
What happens when the selector misses the one token that matters? Any fallback
40 Likes

@Den_zel21
I agree, it’s about frequency, not perfection.
5 Likes

@st0yanov
A few weeks ago I was in a discussion under an X post about LLMs, their adoption and cost. Most people argued it doesn't scale and brought up the usual economic concerns. My take was that it's only a matter of time before the next breakthrough. It's always been this way with tech. Congrats on this one 👏
93 Likes



























