nbardy

7 ms·

1.
by nbardy 12d ago
You can estimate the model size by looking at tokens per second and comparing to open source models
2.
by nbardy 16d ago
One of the amazing things is that when every has one GPU, they will actually have 1k-10k agents at their disposal. LLMs and KV caches have amazing performance characteristics with concurrent throughput. It scales very non linearly. So the t
3.
by nbardy 21d ago
They had incredible revenue growth the last few years and just broke 100M in revenue. I don't know what their internal spend was , but that was almost half of their recent round in ARR. Mostly likely they were profitable or on a clear
4.
by nbardy 24d ago
This is a wildly incorrect and myopic view on the world. Finetuning model is cheap and incredibly useful for deployment. You don't need to pre-train a frontier llm from scratch to make useful models. There is tons of domains where you
5.
by nbardy 24d ago
The real revolution is both. The cost and capability of frontier intelligence will go up AND the cost of "good enough" intelligence will go down.
6.
by nbardy 24d ago
I think this is largely true. Theoretically it follows that mature industries would become this way more and more. In my software engineering jobs there was always strong top down product direction. In my AI research jobs people are often s
7.
by nbardy 1mo ago
I think a lot of it is just time. The quality of a model is E * C Where: E = Efficiency, and efficiency gains come from quality of data, quality of algorithms. C = Compute (Size of model, flops of train run) So a better company can train a
8.
by nbardy 1mo ago
Everyone keeps repeating this who doesn’t understand the underlying technology. Small llms are still way more efficiently server on big GPUs. Sharing server capacity takes advantage of the massive parallel throughput and sharing of memory b
9.
by nbardy 1mo ago
he got promoted to chief scientist and them being an expert at all modeling will really help compared to being llm only
10.
by nbardy 1mo ago
yea, a lot of my prior work was in the "code is the easy part" I was a frontend engineer for years. And something like 90% of my job the code was not the hard part. I loved writing GPU shaders or optimizing visualization performan
11.
by nbardy 2mo ago
The faces look amazing. They can post train for identity preservation easy, that is just an adapter
12.
by nbardy 2mo ago
Then they’re legitimately getting better at svg which is a valuable tool.
13.
by nbardy 2mo ago
This is a weird number. And reeks of the same sort of reallocation fallacy that makes people think the rich making too much makes them poor. If we really just say 4xd everyone’s salary in America. Prices are gonna rapidly rise in everything
14.
by nbardy 2mo ago
These are true and it does make theses fields the first to fall, but also the hype comes form the fact that it can escape these conditions as well: - generalize to non verifiable domains ( https://arxiv.org/abs/2507.1774
15.
by nbardy 2mo ago
We’re definitely going to need a lot of Gpu’s
16.
Shader Benchmark for LLMs(nbardy.github.io)
1 points by nbardy 3mo ago|0 comments
17.
by nbardy 3mo ago
Weird flex. There is cheaper better and faster models you could of moved to with an hour effort
18.
by nbardy 3mo ago
The first bit was interesting and then you flipped right to generic cynicism. They would be impressed with our technology even if it has downsides. Wisdom is knowing humans and technology and imperfect tools.
19.
by nbardy 3mo ago
Signing this sounds like a good way to get fired. Executive in corporations gets to make the decisions. Employment is at will, if you don’t like it you get to leave otherwise you’re not fulfilling your contract
20.
by nbardy 3mo ago
No, look a Composoer 2, it stands out starkly on its own in the pareto frontier on low cast and fast models. Composer 2.5 was a huge leap with minimal compute from xAI. They can compete with OpenAI and anthropic with xAI scale compute. They
21.
by nbardy 3mo ago
It’s a bit misleading to say nothing special, as they are doing more than just increasing parameter count. Progress has been steady in all the sub components of training from data filtering and weighting to sparse attention, optimizers to u
22.
by nbardy 3mo ago
This does feel like the perfect setup for Claude though. Much easier to create a vm testing swarm of 100 disitributions with llms
23.
by nbardy 3mo ago
This has been my thought for a long time. I think all that matters from attention is that there is crosswise comparison going on. You need some amount of parallel compute and some amount of global comparison. And the rest is basically a way
24.
by nbardy 3mo ago
In general this is the way I see open source going. We won't reuse open source libraries as libraries we import, but as design inspiration for the bespoke tools we make. It's too cheap to make your own stuff and too expensive to b
25.
by nbardy 4mo ago
Confidently yes. OpenAI for sure has been training larger models internally and distilling. Pre-training scaling laws all support larger models being more cost effeceint to train then smaller models. And distillation is comparably cheap. So
26.
by nbardy 4mo ago
There is endless returns to frontier intelligence, just because most people can't make use of it doesn't mean someone can't make a ton of money off of it. Most software engineers will just need cheap tokens. But things like p
27.
by nbardy 4mo ago
There is endless returns to frontier intelligence, just because most people can't make use of it doesn't mean someone can't make a ton of money off of it. Most software engineers will just need cheap tokens. But things like p
28.
by nbardy 4mo ago
Seems like wrapping async await functions with CSP was a better way to handle this . Clojure already had a nicer pattern for this
29.
by nbardy 5mo ago
People keep saying this and they don't understand how businesses work. Cursor has 1B in enterprise revenue. It doesn't matter if people can clone their product, those deals don't move slowly
30.
by nbardy 5mo ago
There is step changes that actually merit this though. And a zero day machine IS one of those. It went from 4% zero day success rate to 85% on firefox. Can you not see the significance of that?

More ›