TheCornCollector

joined 6 months ago
[–] TheCornCollector@piefed.zip 5 points 1 week ago

All the sources are in the body of the post. Windows being in free fall is at least questionable to me, because in the source, StatCounter, windows usage has mostly been replaced by ‘unknown’.

[–] TheCornCollector@piefed.zip 11 points 2 weeks ago

They claim it is an open-source model but at least for now it has a custom unspecified license. I hope they change that before the actual release because right now I wouldn’t even call it open-weight

[–] TheCornCollector@piefed.zip 12 points 1 month ago* (last edited 1 month ago) (2 children)

Found it by looking up dark mode Firewatch wallpapers.

Edit: Didn’t find higher resolutions of this specific one. But here are slightly different but higher resolutions variants: https://imgur.com/a/jvkoP

And the second one in this list

[–] TheCornCollector@piefed.zip 0 points 2 months ago* (last edited 2 months ago)

I’m not defending the meme at all. I agreed with the point immediately in the first sentence. I just interpreted the tone of some remarks to be a bit condescending and gave a recommendation based on my opinion on that. Do with it what you want. I’m not here to force any view onto you.

[–] TheCornCollector@piefed.zip 0 points 2 months ago (3 children)

I agree with your first point and I really dislike the amount of misinformation flowing around on both sides, but spending even a little time online will show that is just humans being humans on any topic.

Since any other reference probably wouldn't work here, have one from one of those holistic medicine places

I’m not sure if I’m reading it right, but are you saying that people here, so vegans, don’t trust normal scientific evidence and are only here to reinforce their beliefs by excluding proper evidence? That sounds kind of patronising and implies people are only vegan because of their biased world view instead of proper ethics and science.

[…] to promote a hobby.

This also sounds dismissive of veganism and rude towards people that try their best to reduce harm in a part of their life. Vegans do it because they see it as necessary to live ethically consistent. This is not something done for ✨fun✨.

I would recommend leaving these unnecessary remarks out of your comments if you want to address misinformation such as in this post more effectively while also coming off more respectful.

[–] TheCornCollector@piefed.zip 0 points 2 months ago

https://unsloth.ai/docs/models/qwen3.6#mtp-guide
Unsloth made a guide and has graphs with comparisons

 

Qwen3.6-27B-MTP-UD-Q5_K_XL on my 7900XTX goes from 32 t/s to 50-72 t/s depending on the predictability of the task. So, a 1.5x increase on creative tasks up to a 2.2x increase on math.

MTP does not change the quality with the only cost being a few hundred MB extra VRAM usage. You will need to download a gguf model with MTP support to use it.
My parameters:

; Context memory usage  
ctx-size = 65536  
ctk = q8_0  
ctv = q8_0  

; Prompt processing speed  
batch-size = 1024  
ubatch-size = 1024  

; Speculative decoding  
np = 1  
spec-type = draft-mtp  
spec-draft-n-max = 3  

Edit: did some more testing using Unsloth's parameters and with spec-draft-n-max = 6 I can get up to 82 tk/s, a 2.56x increase, on the same math prompt. But this comes at the cost of the creative writing task that now falls below 40 tk/s.
It seems like this should be tweaked depending on the prompt similar to the sampling parameters.

[–] TheCornCollector@piefed.zip 20 points 2 months ago (1 children)

You can also contribute to OpenStreetMap in your area using simple apps like StreetComplete or EveryDoor. This has a way lower barrier to entry than contributing code in my opinion. And it has the immediate benefit of a better local map for a LOT of services that are built on top of OSM.

[–] TheCornCollector@piefed.zip 9 points 2 months ago

I’m really not fond of the profiling by automated means, but it seems like an inevitable consequence of the design of the threadiverse. Everything is public and easily accessible by anyone that would like to profile you.

I certainly disapprove of moderation based on ideology. Moderation should be based on quality of the content and if it fits in the publicly readable rules. Definitely not some hidden analytics or if the user completely fits in the in-group of the moderator.

I will admit that this might be a good way to find and filter out LLM based bots that are only there to promote or manipulate the conversation. But it should still be done according to public rules.

[–] TheCornCollector@piefed.zip 0 points 2 months ago

Is this post written by an LLM?

[–] TheCornCollector@piefed.zip 3 points 2 months ago (1 children)

I’m no expert, but basically the way to unlock higher/full bandwidth for HDMI 2.1. This will allow the use of higher refresh rate, resolution, and bit depth + HDR. Right now you need to make sacrifices in at least one category with HDMI

[–] TheCornCollector@piefed.zip 22 points 2 months ago* (last edited 2 months ago)

What is the difference between this implementation and the reverse engineered patches that were published a few months ago by Michał Kopeć and Tomasz Pakuła?

Edit: apparently it’s not the same patch, but Tomasz was CC’ed in the patch set so the timing might not be accidental.

[–] TheCornCollector@piefed.zip 18 points 3 months ago* (last edited 3 months ago) (1 children)

I’m European and had to do the same, so it’s based on something else.

 

DeepSeek-V4-Pro-Max, the maximum reasoning effort mode of DeepSeek-V4-Pro, significantly advances the knowledge capabilities of open-source models, firmly establishing itself as the best open-source model available today. It achieves top-tier performance in coding benchmarks and significantly bridges the gap with leading closed-source models on reasoning and agentic tasks. Meanwhile, DeepSeek-V4-Flash-Max achieves comparable reasoning performance to the Pro version when given a larger thinking budget, though its smaller parameter scale naturally places it slightly behind on pure knowledge tasks and the most complex agentic workflows.

Both models support one million tokens context length.


1
Qwen3.6 27B released (huggingface.co)
submitted 3 months ago* (last edited 3 months ago) by TheCornCollector@piefed.zip to c/localllama@sh.itjust.works
 

Recently made a post about the 35b MOE. Now the dense 27b variant has been released.


1
submitted 3 months ago* (last edited 3 months ago) by TheCornCollector@piefed.zip to c/localllama@sh.itjust.works
 

The Qwen3.5 models are still the best local models I’ve used, so I’m excited to see how this updated version performs.

1
submitted 5 months ago* (last edited 5 months ago) by TheCornCollector@piefed.zip to c/freegames@feddit.uk
 

Also available on mobile: Checkout link Android & iOS

 

Long but interesting watch. They talk about his history as a ‘militant vegan’, his current views on effective activism, and if he has any hope left for the future of veganism after he retired.

1
[Steam] Kiki (store.steampowered.com)
 

cross-posted from: https://piefed.zip/c/fosai/p/958141/30b-a3b-glm-4-7-flash-released

Small/fast model with MIT license for local use.

Benchmarks look good for the size. But IMO these smaller models aren’t consistent enough to live up to their promises.

 

Small/fast model with MIT license for local use.

Benchmarks look good for the size. But IMO these smaller models aren’t consistent enough to live up to their promises.

view more: next ›