ilaksh

10 ms·

151.
by ilaksh 4mo ago
"Agents" or "harnesses". We are already there. There are many already. I have my own that I started working on like a year ago.
152.
by ilaksh 4mo ago
Telegram makes it very easy to create a bot. Other systems are either inconvenient or actively hostile to new bots, especially if you're not a well-known business.
153.
by ilaksh 4mo ago
Is there anything we can download? Did they test GLM 5.1?
154.
by ilaksh 4mo ago
Notice a lot of the article was about completions, which fish shell does better than zsh.
155.
by ilaksh 4mo ago
Although I see huge utility in AI, I think he is right in terms of overspending and overenthusiastic build out. Because of for example what Apple is doing by putting an extremely efficient model with task adapters right onto phones. Also be
156.
by ilaksh 4mo ago
I would say about 35% of the time I run into problems and eventually give up and go to GPT 5.5 and it much more efficiently handles the original task. Then I see the token costs going up and it motivates me to continue trying the open sourc
157.
by ilaksh 4mo ago
I think these type of demo videos should allow people to get a sense of super intelligence. Because it's very hard to imagine something that is say three times as smart as you -- by definition you wouldn't be able to comprehend it
158.
by ilaksh 4mo ago
It uses significantly more resources obviously. And/or they have to configure or reconfigure servers for it, which takes time, and doesn't make sense until they have proven the demand at the higher price point.
159.
by ilaksh 4mo ago
The exponential is leading to full compute-in-memory within a few years which will be 100 times more efficient. Which means at least 10 times larger models that are much smarter in addition to extremely fast. It's going to skip the cod
160.
by ilaksh 4mo ago
I'm kind of poor so I have been trying to use DeepSeek v4 Flash, GLM 5.1 etc. as much as possible recently instead of Claude or GPT.
161.
by ilaksh 4mo ago
Use Claude fast mode and turn off thinking. Tell it to just explain what it's plan is to you at a high level. It will go much faster.
162.
by ilaksh 4mo ago
This almost seems like a straw man to me. Isn't the much larger problem the actual increased energy usage and making sure that all of this massive extra cost doesn't just get dumped on consumers? I am a huge proponent of AI actual
163.
by ilaksh 4mo ago
Change the temperature to 0 and it will be more consistent.
164.
by ilaksh 4mo ago
I mean basically you and I are effectively living in parallel universes. Waymo has been running for years, and there are other services including in China and Tesla which is not 100% there but actually very effective. And the thing he compl
165.
by ilaksh 4mo ago
A web search tool to pull up the law that is relevant?
166.
by ilaksh 4mo ago
It can make mistakes and will sometimes, but what he specifically mentioned was a case where it did not pull up a reference that it needed. So using a web search tool effectively would make a big difference.
167.
by ilaksh 4mo ago
3 years max. Maybe 5 if you are lucky.The models will continue to improve. The exponential gains in compute efficiency that have been ongoing for 70+ years will continue and that will result in even smarter models. There are dramatic hardwa
168.
by ilaksh 4mo ago
Jobs have always been a bad deal, especially for most people. Very unfair. Less unfair is entrepreneurship. LLMs (VLMs) etc. should make that more feasible for a broader range of people.
169.
by ilaksh 4mo ago
The article makes it unclear if they are building a new model or if it's just the API. But I am guessing it's the API. So it's "release to developers" rather than "new AI model". They cannot ship the API.
170.
by ilaksh 4mo ago
HN probably has as much as 5 million monthly users. This is not just a small group of insiders, but more of a broadly representative sample of startup and engineering people. So there is a wide range of judgement, and more importantly, a di
171.
by ilaksh 4mo ago
I assume PewDiePie runs something like DeepSeek 4 Flash on that rig.
172.
by ilaksh 4mo ago
Yes Google. They just released their Gemma 4 12b quant.
173.
by ilaksh 4mo ago
OpenAI already had GPT prior to the ChatGPT launch, and I had not really taken it seriously. But on November 30, 2022 when ChatGPT came out and was immediately popular, I reevaluated it. I immediately realized that it meant my time as a pr
174.
by ilaksh 4mo ago
I think that's probably true for the vast majority of Android phones. But if you have a SOTA expensive beast, I wonder if Gemma 4 12B at 4 bit could work? Maybe something like a Redmagic 11 pro or OnePlus 13 running NanoClaw? But also
175.
by ilaksh 4mo ago
But the real bottleneck is the hardware efficiency and not even Karpathy can set up a loop that overcomes that in software. We need the truly compute-in-memory hardware paradigms to be matured and scaled. So it's like recursive hardwar
176.
by ilaksh 4mo ago
Also look into Mythic AI, and also the recent (ultimately tragic) story of the wurtzite ferroelectric device breakthrough at the University of Michigan.
177.
by ilaksh 5mo ago
We've had these threads every month for years.
178.
by ilaksh 5mo ago
Surely these leading tells will be trained out of models pretty soon, given how well known and overused they are. And it might make the writing slightly worse in a way. But it is quite annoying how often this type of construction is used in
179.
by ilaksh 5mo ago
It goes beyond that. There is inherent classism in this, because it implies that you do not question the value of wealthy people who put in relatively little actual work output due to their privileged position. Take for example the unemploy
180.
by ilaksh 5mo ago
All that is really required to prevent the rise of AI and humanoid robot technology from being used to transition from mass exploitation to mass starvation is to stop being so racist and classist and treat everyone fairly. Then we can raise

More ›