asasidh

5 ms·

31.
by asasidh 1y ago
The lessons of Operation Paperclip vs. Operation FIAT offer a surprising historical blueprint in our quest for AGI.
32.
by asasidh 2y ago
thanks for the feedback, will check and update if there are any bugs causing multiple calls
33.
by asasidh 2y ago
agree, the model assumes a multitasking setup where you need some leftover ram for other tasks. You can squeeze in much larger models when running dedicated
34.
by asasidh 2y ago
thanks for the feedback thats a good idea
35.
by asasidh 2y ago
I have Mac going upto 512 GB of memory since the latest mac studio lauched last week has support for 512 GB of unified memory.
36.
by asasidh 2y ago
thanks for the feedback I need to refresh the list or load dynamically.
37.
by asasidh 2y ago
no, I dont see how your post is related. maybe I am just arguing with a bot ?
38.
by asasidh 2y ago
thanks for your feedback, there is room to show how fast or slow the model will run. I will try to update the app
39.
by asasidh 2y ago
yes I agree that you can run. I have personally run Ollama on a 2020 intel macbook pro. Its not a problem of vibe coding, but of the choice of logic i went with.
40.
by asasidh 2y ago
A lot of things can be true at the same time: 1. A lot of scammers launch crypto scams, rug pull tokens, phish/scam people out of their money. 2. Bitcoin is a hard, censorship resistant, permission less money. 3. Outside of Bitcoin, Cr
41.
Show HN: Can I run this LLM? (locally)(can-i-run-this-llm-blue.vercel.app)
42 points by asasidh 2y ago|52 comments
42.
by asasidh 2y ago
yes. The research that goes into it helps be finally learn the concept better. Having a few subscribers also want to spend time to write something and put it out.
43.
by asasidh 2y ago
search, don't sort
44.
by asasidh 2y ago
From custom AI chips to innovative architectures, the battle for efficiency, speed, and dominance is on. But the real game-changer ? Inference compute is becoming more critical than ever—and one company is making serious waves. Groq is emer
45.
by asasidh 2y ago
I would not call this a new insight. BPO is the lowest hanging fruit promised from the AI tree.
46.
Unbundling the BPO: How AI Will Disrupt Outsourced Work(a16z.com)
1 points by asasidh 2y ago|1 comments
47.
by asasidh 2y ago
you can run 32B and even 70B (a bit slow) models on a m4 mac mini pro with 48 GB ram. Out of the box using Ollama. If you enjoy putting together a desktop, thats understandable. https://deepgains.substack.com/p/running-
48.
Ask HN: Will source code cease to be IP?
4 points by asasidh 2y ago|3 comments
49.
Jailbreaking the censorship of DeepSeek R1 locally(deepgains.substack.com)
1 points by asasidh 2y ago|0 comments
50.
Role open for AI Agent only(ycombinator.com)
2 points by asasidh 2y ago|0 comments
51.
by asasidh 2y ago
every one of these products starts out as a fair platform... just give it time and success
52.
Government Wiretaps in U.S. Internet Providers Infiltrated by Chinese Hackers(vulnu.com)
7 points by asasidh 2y ago|0 comments
53.
by asasidh 3y ago
My 2020 Macbook Pro has a function row - touchbar with a "button" for siri.
54.
by asasidh 3y ago
Never used the physical card anywhere. Use it with Apple Pay in B&M as well as web.
55.
by asasidh 3y ago
Would a solution like mixing address the issue with acceptable tradeoffs ? https://arxiv.org/abs/1905.12264
56.
Ask HN: Which are the best newsletters to keep informed on AI development?
49 points by asasidh 3y ago|19 comments
57.
by asasidh 3y ago
It is difficult to get a man or woman to understand something, when their salary depends on them not understanding it.
58.
by asasidh 4y ago
Prompts are kind of like SQL in the structured database world
59.
Ask HN: What do you call technical debt, but for your career?
31 points by asasidh 4y ago|35 comments
60.
by asasidh 4y ago
Finally SEC is getting its act together. Probably triggered by FTX fiasco. Good to see SEC go after the larger players for a change rather than no name exchanges.

More ›