Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
cmogni1
searching Neon…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
28 ms
·
1.
▲
by
cmogni1
3mo ago
I don't mean to be shady, but there are plenty of details that they did release that show that they don't know what they're doing. They make comparisons to FlashAttention-2 when FlashAttention-4 has been out (even if they w
2.
▲
by
cmogni1
3mo ago
Ahh cf my comment above. The cost of failure at scale is too high for a major to just take a new architecture/mechanism and implement it, especially because a) most claims papers make aren't rigorously tested and b) plenty of thin
3.
▲
by
cmogni1
3mo ago
I don't think it makes sense from a business perspective to hold off on details as a new lab. OpenAI will not implement new architectural changes unless they've tested the changes themselves internally. Even if someone claims some
4.
▲
by
cmogni1
3mo ago
I don’t understand why this lab is allergic to providing details on what they actually made, especially when Chinese labs are more than willing to share architectural specs/code/kernels (eg NSA/FSA, RAMBa, HISA, DSA Lightning
5.
▲
Anthropic is pausing the Claude Agent SDK credit change
(twitter.com)
2 points
by
cmogni1
3mo ago
|
0 comments
6.
▲
Kimi K2.7 Code
(platform.kimi.ai)
4 points
by
cmogni1
3mo ago
|
1 comments
7.
▲
DashAttention: Differentiable and Adaptable Sparse Hierarchical Attention
(arxiv.org)
9 points
by
cmogni1
4mo ago
|
0 comments
8.
▲
Anthropic is proof that SaaS isn't dead
(octavehq.com)
1 points
by
cmogni1
4mo ago
|
0 comments
9.
▲
Raven: Memory as a Set of Slots
(goombalab.github.io)
1 points
by
cmogni1
4mo ago
|
0 comments
10.
▲
by
cmogni1
6mo ago
Well I told the ants they were beautiful and they responded with "ty we blushing". I need more Ant Chat in my life :D
11.
▲
The Ballad of Dario and Pete
(twitter.com)
2 points
by
cmogni1
7mo ago
|
0 comments
12.
▲
by
cmogni1
10mo ago
No sadly not :(
13.
▲
Supabase CEO on the "painful" decisions that built a $5B company
(techcrunch.com)
3 points
by
cmogni1
10mo ago
|
2 comments
14.
▲
GTM Engineering Has a Context Problem
(octavehq.com)
7 points
by
cmogni1
10mo ago
|
1 comments
15.
▲
Towards Logic: The Language of AI
(arxiv.org)
3 points
by
cmogni1
11mo ago
|
0 comments
16.
▲
The Future (and Present) of AI Is Synthetic Data
(sutro.sh)
4 points
by
cmogni1
1y ago
|
1 comments
17.
▲
In Defense of the Amyloid Hypothesis
(astralcodexten.com)
4 points
by
cmogni1
1y ago
|
2 comments
18.
▲
OpenAI Seeks Additional Capital from Its Investors as Part of Its $40B Round
(wired.com)
5 points
by
cmogni1
1y ago
|
0 comments
19.
▲
by
cmogni1
1y ago
The article does a great job of highlighting the core disconnect in the LLM API economy: linear pricing for a service with non-linear, quadratic compute costs. The traffic analogy is an excellent framing. One addition: the O(n^2) compute co
20.
▲
No Need for Speed: Why Batch LLM Inference Is Often the Smarter Choice
(sutro.sh)
4 points
by
cmogni1
1y ago
|
0 comments
21.
▲
by
cmogni1
1y ago
Does anyone know how AI coding fits in with S174? If a person’s “coding” part of the job is primarily running prompts and checking code outputs (quality control and minor reprompting) with the remainder of the time used for other activities
22.
▲
Workhorse LLMs: Why Open Source Models Dominate Closed Source for Batch Tasks
(sutro.sh)
118 points
by
cmogni1
1y ago
|
36 comments
23.
▲
System Prompts and Models of AI Tools
(github.com)
1 points
by
cmogni1
1y ago
|
0 comments
24.
▲
by
cmogni1
1y ago
I think the most interesting thing to me is they have multi-hop search & query refinement built in based on prior context/searches. I'm curious how well this works. I've built a lot of LLM applications with web browsing i
25.
▲
Web search on the Anthropic API
(anthropic.com)
272 points
by
cmogni1
1y ago
|
63 comments
26.
▲
Decision Tree (2011)
(mike-naylor.blogspot.com)
1 points
by
cmogni1
1y ago
|
0 comments
27.
▲
Business, Casual (2005)
(thecrimson.com)
2 points
by
cmogni1
1y ago
|
0 comments
28.
▲
Several Psychiatric Disorders Share the Same Root Cause
(sciencealert.com)
8 points
by
cmogni1
2y ago
|
0 comments
29.
▲
Chef – Make programs look like recipes (2003)
(dangermouse.net)
1 points
by
cmogni1
3y ago
|
0 comments
30.
▲
Minerva: A Natural Language Processing (NLP) Model That Solves Math Questions
(marktechpost.com)
25 points
by
cmogni1
4y ago
|
1 comments
More ›