meander_water

8 ms·

91.
by meander_water 10mo ago
I'd like someone to do a comparison of tech company valuations pre GenAI vs post for the same vertical. I understand there's always some optimism for new tech, but the valuations we're seeing seems absurd to me. Like, do they
92.
TurboDiffusion: 100–200× Acceleration for Video Diffusion Models(github.com)
248 points by meander_water 10mo ago|46 comments
93.
by meander_water 10mo ago
Cool project, but loading "Crime and Punishment" crashed my mobile browser. I don't think urls were built for that kind of punishment.
94.
by meander_water 10mo ago
This is actually bearable compared to the new terminal suggestions in vscode. Not only does it autosuggest bizzare completions for commands, it breaks shell completions. So when I tab a file path, it shoves the absolute path into the partia
95.
Tiled Art(tiled.art)
261 points by meander_water 10mo ago|12 comments
96.
Skip the fork – patching code with hatch metadata hooks(vivis.dev)
1 points by meander_water 10mo ago|0 comments
97.
by meander_water 10mo ago
Oh interesting, can you share any links or books?
98.
by meander_water 10mo ago
For anyone interested in the history of IQ tests and how broken they are check out The Mismeasure of Man [0] [0] https://en.wikipedia.org/wiki/The_Mismeasure_of_Man
99.
by meander_water 10mo ago
One could argue that learned speed has the hours of practice "baked in" so it's actually much slower. And that's not a bad thing IMO. I think this post only covers one side of the coin. Sure, getting things done fast ach
100.
by meander_water 10mo ago
This provides a bit more detail on how they relate to each other https://www.copilotkit.ai/ag-ui-and-a2ui
101.
by meander_water 11mo ago
Overall really interesting read, but I'm having trouble processing this: > OpenRouter performs internal categorization on a random sample comprising approximately 0.25% of all prompts How can you arrive at any conclusion with such a
102.
by meander_water 11mo ago
Yeah I was hoping for a lot more from the title.
103.
by meander_water 11mo ago
This is really cool and interesting timing, as I created something similar recently - https://github.com/julio-mcdulio/pmp I've been using mlflow to store my prompts, but wanted something lightweight on the cli to
104.
by meander_water 11mo ago
We already have AG-UI [0], which has been implemented by frameworks like Microsoft agent framework, pydantic AI and llamaindex. I guess they'll just have to duplicate functionality. Sigh. [0] https://docs.ag-ui.com/intr
105.
by meander_water 11mo ago
I've been doing something like this for a while, and it provides decent enough results for agents to one-shot. The key is to specify the LLM with the idea, and then get it to ask you questions until you feel enough ambiguity has been e
106.
by meander_water 11mo ago
It's funny, I clicked the link to the demo, but it 404s, then I tried googling Worldgen, and it turns out someone else has built the same thing in May and called it Worldgen as well. Looks like it does better at realistic 3D scenes com
107.
Open source image generation with style codes (–sref))(kwai-kolors.github.io)
1 points by meander_water 11mo ago|0 comments
108.
A Style is Worth One Code: open-source Midjourey-like --sref(arxiv.org)
2 points by meander_water 11mo ago|0 comments
109.
Solving a Million-Step LLM Task with Zero Errors(arxiv.org)
2 points by meander_water 11mo ago|1 comments
110.
by meander_water 1y ago
Here's the juicy stuff mentioned in the article: https://storage.courtlistener.com/recap/gov.uscourts.cand.43...
111.
by meander_water 1y ago
I've noticed this too. The latest cursor version has a @browser command which will launch a browser with playwright, and call tools to inspect the html and inject JavaScript to debug in real-time. When it has access to the right tools,
112.
Tongyi DeepResearch – open-source 30B MoE Model that rivals OpenAI DeepResearch(tongyi-agent.github.io)
365 points by meander_water 1y ago|153 comments
113.
by meander_water 1y ago
I think you can compile the rust core lib from source yourself - https://github.com/Shyam20001/brahma-core
114.
by meander_water 1y ago
I don't get it, can't you just use yaml instead of inventing another DSL.
115.
by meander_water 1y ago
There's a few bits of information from the original sources that's left out: - The METR paper surveyed just 16 developers to arrive at their conclusion. Not sure how that got past review. [0] - The finding from the MIT report can
116.
Unicode Footguns in Python(pythonkoans.substack.com)
43 points by meander_water 1y ago|20 comments
117.
by meander_water 1y ago
> There is a user in the JavaScript community who goes around adding "backwards compatibility" to projects. They do this by adding 50 extra package dependencies to your project, which are maintained by them. This is a spicy one
118.
by meander_water 1y ago
Yeah I was hoping to see the actual script content
119.
by meander_water 1y ago
Why wouldn't you use OP's approach to build up the representative embeddings, and then train the MLP on that? That way you can effectively handle open sets and train a more accurate MLP model. With your approach I don't think
120.
by meander_water 1y ago
Agreed, I've run sentence-transformers/all-MiniLM-L6-v2 locally on CPU for a similar task, and it was approx X2 faster than calling the OpenAI embedding API, not to mention free.

More ›