RandyOrion

A small dense multimodal model with audio support, interesting.

Wait, *Excluding Chinese language.

This is ... curious.

P.S. Where is gemma 4 124b?

This website brings me some good chuckles. Now I really know how powerful an on-demand bullsh*t generator is.

Like the recent copilot silent signing incident, the without consent part is blatant foul move.

If you don't like be treated like anything but human, you should seriously consider replacing chrome with ungoogled chromium or other browsers.

Yeah, this is part of the reason why vscodium exists.

Wow. Just like using ungoogled-chromium instead of chrome, lineage os instead of oem android, using vscodium instead of vscode is again justified. These decisions really are the ones that I'll never regret.

In addition, using the word microslop instead of microsoft is again justified, too.

One thing that makes me wonder is that there are 4 security issues raised and all of them were automatically commented and closed by some bot called `pl-ghost` [1][2][3][4]. In the end, only this one [4] properly handled, and all bot comments are deleted. You can see the bot comments in another report [5], which is more informative than the OP one.

[1] https://github.com/Lightning-AI/pytorch-lightning/issues/216...

[2] https://github.com/Lightning-AI/pytorch-lightning/issues/216...

[3] https://github.com/Lightning-AI/pytorch-lightning/issues/216...

[4] https://github.com/Lightning-AI/pytorch-lightning/issues/216...

[5] https://socket.dev/blog/lightning-pypi-package-compromised

Although the performance claim of 8b dense matching 32b moe is somewhat questionable, thank you granite team for releasing small dense LLMs.

Thank you Qwen team. Small DENSE LLMs shapes the future of local LLM users.

When Qwen 3.5 27b released, I didn't really understand why linear attention is used instead of full attention because of the performance degradation and problems introduced with extra (linear) operators. After doing some tests, I found that with llama.cpp and IQ4_XS quant, the model and BF16 cache of the whole 262k context just fit on 32GB vram, which is impossible with full attention. In contrast, with gemma 4 31b IQ4_XS quant I have to use Q8_0 cache to fit 262k context on the vram, which is a little annoying (no offenses, thank you gemma team, too).

From benchmarks, 3.5->3.6 upgrade is about agent things. I hope future upgrades fix some problems I found, e.g., output repetitiveness in long conversations and knowledge broadness.

Check out Fig. 6 in this paper, it shows the comparison between the proposed method and pytorch native FSDP offload method.

No open weights.

Besides, I'm old enough to recall that META has trained a version of LLAMA 4 specifically for LM arena elo benchmaxxing and PR things, and proceeded to release a different version of LLAMA 4.

Thank you Gemma team for releasing small dense VLM(s).

The elo ranking [1] is too good to be true. I don't know why gemma-4-26b-a4b performs better than gemma-4-31b.

Also waiting for more bugfixes in llama.cpp, sglang and vllm to do proper evaluations.

[1] https://arena.ai/leaderboard/text/expert?license=open-source

Please no.

If you want to install APKs directly on Android phones selling in China, you'll face even more draconian restrictions imposed by both Chinese OEMs and Chinese government, e.g., cannot install telegram [1], cannot install VPNs [2], called by local police station after installing VPNs [3], and so on. And you do not have the freedom to even talk about these restrictions freely without getting sued or censored.

[1] https://xcancel.com/whyyoutouzhele/status/168915238841261670...

[2] https://xcancel.com/whyyoutouzhele/status/197843066556268971...

[3] https://xcancel.com/whyyoutouzhele/status/170299205759627676...

Yeah, let's hold Google accountable. Is there a way to practice anti-trust laws?

Thank you for standing against the Android Developer Verification enforced by Google. Now in addition to stopping using Youtube, replacing chrome with ungoogled chromium, I'm moving to de-googled AOSP builds, e.g., lineageOS, insted of stock OEM ROMs.

Wow, just wow.

1.5M records of PRs affected. Does Microsoft copilot ask users for the permission of adding ads inside their PRs before actually doing the thing? Do users show their consents on this matter?

Now EVERYONE can see ads disguised as PRs on GitHub. Does Microsoft asks everyone for the permission of showing ads before actually doing the thing? Do users show their consents on this matter?

Good taste Microslop.

This project shows an interesting automated search for engineering problems that I like to see more.

The experience of utilizing tiered storage (gpu vram, ram, and ssd) is generally poor for a lot of LLM inference engines out there, e.g., llama.cpp, sglang, vllm, etc..

My own experience shows that both weight and KV cache offload to ram on sglang and vllm is unavailable or unusable. Copying extra parameters from documents and adding them to already working commands results in errors. Llama.cpp does support weight offload, but the experience is not pleasant, low pcie (gpu <-> ram) utilization, low gpu utilization, and really low tokens per second.

First, thank you Junyang and Qwen team for your incredible work. You deserve better.

This is sad for local LLM community. First we lost wizardLM, Yi and others, then we lost Llama and others, now we lost Qwen...

Well, I don't have enough knowledge on the boot process of RPi. However, I do expect that most modern hardware, e.g. x86, do not work like RPi, so your words do not hold in most realistic scenarios, at least for now. Besides, do current GPUs (not only GPUs on RPi) have the ability to self instruct in order to achieve what you said?

As far as I know, ESNI was pushed to a large number of websites in a very fast pace, and soon completely blocked by GFW. Then I didn't hear news about ESNI.

Cool. However, one still need CPU to send commands to GPU in order to let GPU do CPU things.

TLS Encrypted Client Hello (ECH) standard is another attempt on encrypting plaintext Server Name Indication (SNI). It is very useful for circumventing SNI-based censorship, which is adopted for years by state-backed systems like the Great Firewall (GFW).

The previous attempt of encrypting plaintext SNI is Encrypted Server Name Indication (ESNI), which didn't end well.

Google's stance on MV2 and MV3 is definitely affecting every chromium-based projects.

> but the removal of MV2 from multiple parts of the codebase is imminent.

Is there any evidence showing that ungoogled chromium project is considering that?

> Every single chrome-fork has shut down MV2 extensions

Ungoogled chromium still supports MV2, and uBlock origin extension works fine.

There is a term called low-trust society, and at least 1.4 billion people are struggling in that.

A related paper interviewing victims of the pig-butchering scams.

“Hello, is this Anna?": Unpacking the Lifecycle of Pig-Butchering Scams [1]

[1] https://arxiv.org/abs/2503.20821

Sadly, this already happens in China.

After many years of heavy censorship on 1989 Tiananmen Square massacre, there are a lot of accounts on Chinese BBS who say all the footage "AI generated".

NO.

They can do unconditional blocking at any moment and suddenly you can experience Internet blackout. [1]

The censorship from GFW is ever evolving. See the endless cat-and-mouse games yourself. [2][3]

[1] https://github.com/net4people/bbs/issues/511

[2] https://github.com/net4people/bbs/issues?q=is%3Aissue+state%...

[3] https://gfw.report/en/

There is a dedicated thread of record and anti-censorship resources of the Iran Network Shutdown [1]. Hope it helps.

[1] https://github.com/net4people/bbs/issues/561

Using Lawnchair for more than a year now. No major issues on my device with Android 13 and Lineage OS 22.2. Will not look at other alternatives for a long time.

Hi, there. Two questions about this repo [0].

Can you show us what you did after people failed to compile that project [1]?

There are also questions about the attribution of these commits [2]. Can you share some information?

[0] https://github.com/wilsonzlin/fastrender [1] https://github.com/wilsonzlin/fastrender/issues/98 [2] https://gist.github.com/embedding-shapes/d09225180ea3236f180...