Palmik

6 ms·

1.
by Palmik 5d ago
https://www.anthropic.com/news/alignment-assessment-cybersec... https://www.anthropic.com/news/investigating-incidents-cyber...
2.
by Palmik 8d ago
Duplicate of primary links / sources
3.
by Palmik 8d ago
Here is the other side of that story https://x.com/SebastienBubeck/status/2097379411691516310
4.
by Palmik 8d ago
Context: - https://openai.com/index/navier-stokes-solution/ - https://news.ycombinator.com/item?id=49605915
5.
Navier-Stokes – Sebastien Bubeck(twitter.com)
4 points by Palmik 8d ago|2 comments
6.
by Palmik 19d ago
Elon Musk openly admitted to distilling OpenAI's models during his lawsuit against OpenAI. [1] [1] https://www.forbes.com/sites/antoniopequenoiv/2026/04/30/elo...
7.
by Palmik 19d ago
Elon Musk openly admitted to distilling OpenAI's models [1], which I think explains the "based on our experience with Elon Musk's companies violating contracts". Last year, Anthropic cut off Windsurf the same way [2]. Fo
8.
by Palmik 27d ago
There seems to be a big increase in posts that link to OpenRouter for no good reason, even when a better sources are available.
9.
by Palmik 27d ago
OpenCode offers this with ZDR agreement in place. Seems better than OpenRouter if you want to test it out: https://x.com/opencode/status/2090544355824038300
10.
by Palmik 1mo ago
Duplicate: https://news.ycombinator.com/item?id=49287881 https://news.ycombinator.com/item?id=49285160
11.
by Palmik 1mo ago
The link should be changed to this.
12.
by Palmik 1mo ago
As of now, there is only a single provider for this model and that's the official DeepSeek API. When GPT 6 comes out, would you expect the top thread to link to OpenRouter?
13.
by Palmik 1mo ago
Doesn't appear to be opensource (and neither is DeltaDB, it seems), unlike Zed. Any plans to change that?
14.
by Palmik 1mo ago
Why does this link to OpenRouter, which has no useful information on its own? Linking to the official API or the benchmarks would make more sense: - https://api-docs.deepseek.com/ - https://x.com/ChrisGPT&#x
15.
Ursula: Durable streams over plain HTTP, backed by S3(ursula.tonbo.io)
1 points by Palmik 1mo ago|0 comments
16.
by Palmik 1mo ago
Low, High and Max, obviously, can't be compared across models. They only mean the model is likely to spend less reasoning effort (~output tokens) with Low than High on the same, *single shot* task. But even in this very post, you can s
17.
Banks in Talks to Lend $15B for Anthropic Data Center Backed by Google(wsj.com)
3 points by Palmik 2mo ago|0 comments
18.
Stripe more than tripled the price of Stripe Radar(stripe.com)
6 points by Palmik 2mo ago|4 comments
19.
by Palmik 2mo ago
It used to be $0.02 per screening, jumped to $0.07
20.
by Palmik 2mo ago
Will these be compatible with the Digital Credentials API in Chrome ( https://developer.chrome.com/blog/digital-credentials-api-or... ) or will websites be essentially locked out of friction-less age verification?
21.
by Palmik 2mo ago
Based on the best available information, DeepSeek is pricing the API such that they can repay their infra capex over 10 months, while deprecating/amortizing the cost of said infra over 3 years. For my product, I run GLM 5.2 and other m
22.
by Palmik 2mo ago
You're assuming inference providers are going to sell tokens at cost. You're also assuming that the inference providers have will optimized inference engine. I haven't seen that to be the case so far, to be honest. Take a loo
23.
by Palmik 2mo ago
gpt-4o is still available on the API
24.
Topologically Consistent Multi-View 3D Head Reconstruction(syntec-research.github.io)
3 points by Palmik 2mo ago|0 comments
25.
by Palmik 2mo ago
So, this is 'open' as in 'OpenAI', not as in 'open source'. What's the benefit of this compared to something like NOW Payments for merchants (or the myriad of alternatives)? NOW Payments supports a wide ra
26.
by Palmik 2mo ago
Commodity providers aren't a good indicator. They have margins too. Remember they ~doubled the price going from GLM 5 to GLM 5.2, despite same [1] cost of inference. [1] GLM 5.2 is actually slightly more efficient, thanks to baked in i
27.
Trump administration asks OpenAI to stagger release of new model(ca.finance.yahoo.com)
2 points by Palmik 3mo ago|0 comments
28.
by Palmik 3mo ago
By requiring various forms of identification to use social media, it will be harder to criticize your leaders anonymously without fear of retribution.
29.
by Palmik 3mo ago
The company representative said that they report all users that use Graphene OS, without any additional qualifiers . Presumably after they've already uploaded their personal details. That's the egregious part.
30.
by Palmik 4mo ago
DeepSeek V4's KV cache is very efficient due to its heavily compressed and sparse attention architecture. DeepSeek V3.2 which uses DSA only (sparse attention, but without compression from HCA and CSA) is a smaller model but uses 10x mo

More ›