Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
k2xl
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
9 ms
·
91.
▲
by
k2xl
1y ago
> The least-likely part of the work is behind us; the scientific insights that got us to systems like GPT-4 and o3 were hard-won, but will take us very far. > 2026 will likely see the arrival of systems that can figure out novel insig
92.
▲
by
k2xl
1y ago
Not distilled, same model. https://x.com/therealadamg/status/1932534244774957121?s=46&t...
93.
▲
by
k2xl
1y ago
I recommend you try it before judging. I will be honest it has been actually very useful. You are also assuming that the LLM is providing false information (or will have a higher chance of providing false information than a human)
94.
▲
by
k2xl
1y ago
I've done something similar for learning about a controversial topic. I ask it to act as if it is called Bob is a well informed supporter of one side (like Ukraine) and then act as if it is something named Alice who is a well informed
95.
▲
by
k2xl
1y ago
I think i mentioned this in my original reply- the app was already built and too integrated with supabase, it is hard to leave
96.
▲
by
k2xl
1y ago
I've been a lukewarm user of Supabase for my side projects. Unfortunately the amount of work to get off of it has been too high for me to leave. The major issue is - cost. It is way more expensive than I realized as they have so many l
97.
▲
by
k2xl
1y ago
There is a similar issue with image and video generation. Asking the AI to "Generate an image of a man holding a pencil with his left hand" or "Generate a clock showing the time 5 minutes past 6 o'clock" often fail
98.
▲
by
k2xl
2y ago
I can’t use their video gen model veo2 since there is a waitlist… hard to tell if they are winning in video when they haven’t scaled that product.
99.
▲
by
k2xl
2y ago
He isnt saying texas is less bias, but from public perception perspective, california is biased.
100.
▲
by
k2xl
2y ago
Something not mentioned is AI generated music. Suno's development this year is impressive. Unclear what this will mean for music artists over next few years.
101.
▲
by
k2xl
2y ago
What model does it use under the hood? How much context window does it load when it is solving tasks? How does it determine which files to load into context?
102.
▲
by
k2xl
2y ago
It says at the top $500 per month
103.
▲
by
k2xl
2y ago
This is impressive, but why are they all looking still like a video game? Could they have this render movie scenes with realistic looking humans? I wonder if it is due to the training set they use being mostly video games?
104.
▲
by
k2xl
2y ago
Not to dispute the main point you are making, but that's assuming 360K salary. OpenAI engineers make significantly more than that.
105.
▲
by
k2xl
2y ago
Trying to find a pricing page.. Anyone find one?
106.
▲
by
k2xl
2y ago
The amount of data that would be required for a lookup table for all best moves for every board configuration would be infeasible. They have managed to create one for 7 pieces. Last update on trying to get to 8 piece database: https:/
107.
▲
by
k2xl
2y ago
Could o1 have helped the hunters finally reach it?
108.
▲
by
k2xl
2y ago
Pricing page updated for O1 API costs. https://openai.com/api/pricing/ $15.00 / 1M input tokens $60.00 / 1M output tokens For o1 preview Approx 3x the price of gpt4o. o1-mini $3.00 / 1M input tokens
109.
▲
by
k2xl
2y ago
Been hesitant to get due to battery life. My fitbit lasts for days on one charge. Is there ways to configure the watch settings to extend the battery life for more than a day?
110.
▲
by
k2xl
2y ago
that would be great... i'm curious on the cost per million token as it compares to gpt-4o-mini
111.
▲
by
k2xl
2y ago
Nice - btw curious how much it costs to run Llama on this type of side project? I can imagine the cost goes to some reasonable amount when things like this get shared on HN. Does Fly.io give a flat price to use their inference? Curious how
112.
▲
by
k2xl
2y ago
This is great - Though I am confused on two things: 1. How is it possible that GPT-4o mini outperforms 3.5 turbo but 3.5 turbo is more expensive? Like why would someone use a worse model and pay more? 2. Why is the GPT4o vision and GPT4o-mi
113.
▲
by
k2xl
2y ago
Shameless plug for a side project i have been working on for a few years, but if you like Baba is You you may enjoy Pathology or Sokopath. They can be played on my site Thinky.gg ( https://thinky.gg ) Theres a pretty active commun
114.
▲
by
k2xl
2y ago
Would be great if it had Zwave and Zigbee support... Would love this for home automation use cases
115.
▲
by
k2xl
2y ago
Lot of investment in new data centers getting built from big tech companies. I imagine they want to highlight chips that will be ready in time for those new constructions to break ground.
116.
▲
by
k2xl
2y ago
2% conversion for an 8K purchase is actually incredibly high. Bought in 2018 and v12 has been absolutely amazing. Next version 12.4 will also remove the steering wheel nag.
117.
▲
by
k2xl
2y ago
Would be neat to see this on much older videos (maybe WW2 era) to see how it improves details.
118.
▲
by
k2xl
3y ago
Curious on if others have switched from vscode to Zed and can share their experiences
119.
▲
by
k2xl
3y ago
Turbo tax comes to mind
120.
▲
by
k2xl
3y ago
Shameless plug, but if you are interested in Sokoban type games check out https://thinky.gg . Features a fun Sokoban variant (called Sokopath) as well as another NP hard variant called Pathology (goal is to get from point A to p
More ›