daemonologist

8 ms·

91.
by daemonologist 6mo ago
You can chat with the model on the project page: https://indiepixel.de/meful/index.html It (v3) mostly only says hello and bye, but I guess for 25k parameters you can't complain. (I think the rather exuberant cop
92.
by daemonologist 6mo ago
I believe part of the legislation is that manufacturers must make spare parts available for five years.
93.
by daemonologist 6mo ago
I remember talking to some of the folks running UIUC's hackathon (probably ten years ago) and they'd built a sort of page-rank for Github - hand-identifying the most prominent and reputable projects/individuals and then using
94.
by daemonologist 6mo ago
> This puts the US government into a loose / loose position. You might even call it... a tight spot
95.
by daemonologist 6mo ago
AMD has built some consumer GPUs in the recent past with HBM - RX Vega and Radeon VII (although I assume not all "HBM" is created equal).
96.
by daemonologist 6mo ago
Note that these are the "chat" system prompts - although it's not mentioned I would assume that Claude Code gets something significantly different, which might have more language about malware refusal (other coding tools woul
97.
by daemonologist 6mo ago
The benchmark GP mentioned is measuring at 128k-256k context (there's another at 524k-1024k, where 4.6 scored 78.3% and 4.7 scored 32.2%). The longer the context the worse the performance; there isn't really a qualitative step cha
98.
by daemonologist 6mo ago
I didn't see a direct comparison, but there's some overlap in the published benchmarks: │ Qwen 3.6 35B-A3B │ Haiku 4.5 ────────────────────────┼──────────────────┼────────────────────
99.
by daemonologist 6mo ago
The 3B active is small enough that it's decently fast even with experts offloaded to system memory. Any PC with a modern (>=8 GB) GPU and sufficient system memory (at least ~24 GB) will be able to run it okay; I'm pretty happy
100.
by daemonologist 6mo ago
No - this model has the weights memory footprint of a 35B model (you do save a little bit on the KV cache, which will be smaller than the total size suggests). The lower number of active parameters gives you faster inference, including low
101.
by daemonologist 6mo ago
ROCm usually only supports two generations of consumer GPUs, and sometimes the latest generation is slow to gain support. Currently only RDNA 3 and RDNA 4 (RX 7000 and 9000) are supported: https://rocm.docs.amd.com/projects
102.
by daemonologist 6mo ago
Among consumer cards, latest ROCm supports only RDNA 3 and RDNA 4 (RX 7000 and RX 9000 series). Most stuff will run on a slightly older version for now, so you can get away with RDNA 2 (6000 series).
103.
by daemonologist 6mo ago
Good domain name.
104.
by daemonologist 6mo ago
It's often found alongside natural gas because the rock structures that can trap methane can also trap other gasses, but the original source is different - thermal decomposition of organic matter for natural gas and radioactive decay,
105.
by daemonologist 6mo ago
Obvious example is a corporate chatbot (if it's using tools, probably for internal use). Non-technical users might be accessing it from a phone or locked-down corporate device, and you probably don't want to run a CLI in a sandbo
106.
by daemonologist 6mo ago
The idea is the opposite - "nobody" can make money selling software anymore, because software can be cheaply created by an LLM, so you want to start a business that previously would have had to buy software/software enginee
107.
by daemonologist 6mo ago
Whisper is still old reliable - I find that it's less prone to hallucinations than newer models, easier to run (on AMD GPU, via whisper.cpp), and only ~2x slower than parakeet. I even bothered to "port" Parakeet to Nemo-les
108.
by daemonologist 6mo ago
It's interesting to me that all AI music sounds slightly sibilant - like someone taped a sheet of paper to the speaker or covered my head in dry leaves. I know no model is perfect but I'd have thought they'd have ironed out
109.
by daemonologist 6mo ago
If you scroll to the bottom of that page, they discuss possible evidence of damage to the radar from satellite imagery.
110.
by daemonologist 6mo ago
iNaturalist is cool, but it'd be a lot cooler if they released their models.
111.
by daemonologist 6mo ago
The implication is that there is (should be) a major speed difference - naively you'd expect the MoE to be 10x faster and cheaper, which can be pretty relevant on real world tasks.
112.
by daemonologist 6mo ago
Gemma will give you the most, Gemini will give you the best. The former is much smaller and therefore cheaper to run, but less capable. Although I'm not sure whether Gemma will be available even in aistudio - they took the last one do
113.
by daemonologist 6mo ago
It will probably lead to more cars traveling at any time, but potentially far fewer cars parked. However: turn most of the street parking into bus and/or bike lanes, the parking garages into apartments, seems like an absolute win. (
114.
by daemonologist 6mo ago
I'm pretty sure they're referring to their coworker as "he," not an LLM.
115.
by daemonologist 6mo ago
VR's problem, in my opinion, is that I can get immersed (fully, exactly as the author describes it) in a 2D game just fine - the lack of stereo vision or head-tracking or motion controls is no more an impediment to my immersion than th
116.
by daemonologist 6mo ago
If you stick with your OS/package manager-distributed version, installation isn't painful anymore (provided that version approximately overlaps with your generation of GPU). It's okay for inference, and okay for training if
117.
by daemonologist 7mo ago
Yes! After many years of using only linux or windows machines, I was assigned an iMac at an internship and noticed the friction with fullscreening things. I decided not to fight it and spent the next year happily working in little windows
118.
by daemonologist 7mo ago
7800 XT has 624 GB/s as well, and can be found for $400 used. 16 GB of course.
119.
by daemonologist 7mo ago
Had to break out Chromium for this one - Firefox+Linux does not like webgpu (my whole DE started flickering).
120.
by daemonologist 7mo ago
Or the classic from Dijkstra ( https://www.cs.utexas.edu/~EWD/transcriptions/EWD08xx/EWD867... ): > even Alan M. Turing allowed himself to be drawn into the discussion of the question whether computers can t

More ›