scosman

6 ms·

61.
by scosman 4mo ago
They copied strings, and visual design. Both of which are work, both of which are covered by copyright.
62.
by scosman 4mo ago
Short segments of popular works sure. Many UI pages with identical layouts and copy, essentially zero chance. The agent had access to the original code at inference time.
63.
Show HN: Create animated explainer video from a prompt(github.com)
2 points by scosman 4mo ago|0 comments
64.
by scosman 4mo ago
"ghostlygourd.com" is a S+ tier domain. Would click
65.
by scosman 4mo ago
It's not economic to run them locally. It's amazing for privacy, and fun hobby. But you're either looking at super slow CPU builds with $10k in RAM, $90k worth of GPUs, or a really quantized model that doesn't compare in
66.
by scosman 4mo ago
anyone manage to get a point?
67.
by scosman 4mo ago
Yeah totally agree! My pet peeve feature list is also better for society than donating $500M to charity!
68.
by scosman 4mo ago
Top right: link to the Lore Repo on Github
69.
by scosman 4mo ago
My 8yo son plays with this any chance he gets. We've started building watches together as a result (simple "drop in movement mods", but aspirations of building our own movement). The author really made something special.
70.
by scosman 4mo ago
I think the point is: if every app chooses the best model for their use case, my phone is hosed (disk, memory swap, memory). A good-enough default might be better for the user than the each app having the best possible.
71.
by scosman 4mo ago
ahh yes. I'm using Whisper v3 turbo via WhisperKit as well. Will play with parakeet
72.
by scosman 4mo ago
I think that's what they are trying to avoid. If you need on-device intelligence, their pitch was "The model the device already has is best", and if you need something more specific an adapter (aka, a fine-tune/lora) is
73.
by scosman 4mo ago
I had the same experience so started building my own. All problems are solvable, just working on the polish. - crash recovery: part one is use ADTS aac (even if process crashes, audio is saved up until it does). Part two is isolating the tr
74.
by scosman 4mo ago
Those transcription times are fast fast. What model/library do you use?
75.
Kagi adds Hacker News conversation under links(imgur.com)
5 points by scosman 4mo ago|0 comments
76.
by scosman 4mo ago
They actively use the name https://www.war.gov
77.
by scosman 4mo ago
This is likely correct, sorry for the bad info. Was working from memory.
78.
by scosman 4mo ago
Free server-size model access for apps with <2M downloads, getting the same privacy guarantees. Hopefully they scale this up to all apps in time (I assume hardware/cost constrained, but larger devs would pay). https://dev
79.
by scosman 4mo ago
> just a privacy-polished frontend for Gemini I'd use this. I'd rather have strong privacy guarantees, but this is still good.
80.
by scosman 4mo ago
Cerebras is trialing Kimi K2.6 at 3000t/s (invite only). I'm excited for when the fast hardware gets more mainstream for frontier models. Models designed for speed on Nvidia are nice addition that could bridge the gap.
81.
by scosman 4mo ago
Google's QAT claims to need 6.7 GB RAM, vs Unsloth's dynamic quants at 8GB. Would love to see some benchmarks. Both amazing for size.
82.
by scosman 4mo ago
The image on the Nvidia announcement tells the story: 6 identical laptops from various manufacturers running the same SoC. The Microsoft one is just one of many. https://www.nvidia.com/en-us/products/rtx-spark&#x2F
83.
by scosman 4mo ago
They also do a good job working over the little differences between APIs. Tool calling sometimes breaks on major providers, and OR will patch it before the provider does. Libraries like LiteLLM do this too, but OR is faster.
84.
by scosman 4mo ago
until GPT 5.4, Claude always had a decent edge in benchmarks (any date before then). The gap was huge during the Sonnet 3.x vs GPT 4.x eta.
85.
by scosman 4mo ago
Benchmarking 1 or a few samples isn't ever going to yield anything but noise. The actual benchmarks use thousands of tasks. GPT 5.5 genuinely was back on top for a while there, but if you look at the past 2 years, being on Claude was b
86.
by scosman 5mo ago
not your fault, the article uses that language!
87.
by scosman 5mo ago
I would have guessed that was just a bad title here but no, article states it as "opted in by default".
88.
by scosman 5mo ago
even the color match...
89.
by scosman 5mo ago
beautiful. But looks like an Aston Martin
90.
by scosman 5mo ago
yup.

More ›