0xSero's banner

0xSero

@0xSero • 65,414 subscribers

Do you see what I see?

Shorts

100 tok/s on GLM-5.3-Flash, GLM-5.3 runs on the 6000s @ 80-180 tok/s I have not tried the Studio, I’m sure it’s an amazing machine especially if you are in the Mac world already. From a pure logistics POV the sparks use little power, make less noise, serve good model & tokens

100 tok/s on GLM-5.3-Flash, GLM-5.3 runs on the 6000s @ 80-180 tok/s I have not tried the Studio, I’m sure it’s an amazing machine especially if you are in the Mac world already. From a pure logistics POV the sparks use little power, make less noise, serve good model & tokens

67,414 görüntüleme

Tibo, why? Why must the performance suck so bad.

Tibo, why? Why must the performance suck so bad.

121,863 görüntüleme

Deepseek-v4-Flash running locally. Makes me dizzy, this is one of 6 sessions running rn

Deepseek-v4-Flash running locally. Makes me dizzy, this is one of 6 sessions running rn

37,143 görüntüleme

Codex nooooooo I can't close this popup

Codex nooooooo I can't close this popup

135,878 görüntüleme

Kitty litter, GLM-5.2 running locally. Have you seen 180 tok/s? Do you see how much room we have to push things forward? Most owners don’t know how to squeeze every token out of their GPUs

Kitty litter, GLM-5.2 running locally. Have you seen 180 tok/s? Do you see how much room we have to push things forward? Most owners don’t know how to squeeze every token out of their GPUs

22,222 görüntüleme

GLM-5.2-REAP-NVFP4 This experiment I calibrated the model on 30,000+ samples of my agent sessions, tweets, writing, and codebases. Pretty charming, every dataset produced something different.

GLM-5.2-REAP-NVFP4 This experiment I calibrated the model on 30,000+ samples of my agent sessions, tweets, writing, and codebases. Pretty charming, every dataset produced something different.

22,535 görüntüleme

If you've done work in OSS AI, or Local AI and want to have you and your company on State of Local AI - 2026 Get in touch, you'll be on as a coauthor with the companies and people who make this all possible. We moved release date to August 7th, we have 6 contributors so far

If you've done work in OSS AI, or Local AI and want to have you and your company on State of Local AI - 2026 Get in touch, you'll be on as a coauthor with the companies and people who make this all possible. We moved release date to August 7th, we have 6 contributors so far

13,786 görüntüleme

The fact that a 284B parameter model is able to in a single 400K token session - ssh into my Homelab to find docs - ssh into lambda / prime intellect - rewrite reap to support the DS4 attention - download models & datasets - run tests to check it works

The fact that a 284B parameter model is able to in a single 400K token session - ssh into my Homelab to find docs - ssh into lambda / prime intellect - rewrite reap to support the DS4 attention - download models & datasets - run tests to check it works

22,762 görüntüleme

What 4 concurrency on GLM sounds like. So gosh damn loud, open air frames are easier to build but so gosh darn loud

What 4 concurrency on GLM sounds like. So gosh damn loud, open air frames are easier to build but so gosh darn loud

11,809 görüntüleme

Asking Codex (GPU) to find and kill Codex (Cerebras)

Asking Codex (GPU) to find and kill Codex (Cerebras)

24,131 görüntüleme

Videos

Anya Rossi

sweetdream.ai

SweetDream.ai•Sponsored•Livecam

Watch Anya Live

Anya is streaming live right now! Join her private show and enjoy exclusive content.

Exclusive private shows

1.2k viewers online

Private Show

Join now for exclusive access

Free preview available • Premium content

Opus at home, at 200+ tok/s I love ZAI

Opus at home, at 200+ tok/s I love ZAI

200,857 görüntüleme • 7 gün önce

Local AI will be right in your OS Omarchy, PR up soon.

Local AI will be right in your OS Omarchy, PR up soon.

54,800 görüntüleme • 5 gün önce

I just saw AGI, what the heck. Deepseek-v4-flash-0731 invented vision for itself, this is so incredible. Result in the next post:

I just saw AGI, what the heck. Deepseek-v4-flash-0731 invented vision for itself, this is so incredible. Result in the next post:

174,038 görüntüleme • 21 gün önce

Yours to keep.

Yours to keep.

144,267 görüntüleme • 29 gün önce

I can’t believe someone just gave me 4 RTX pro 6000s, it’s absolutely one of the greatest acts of generosity someone has shown me. The money I would have spent on the cards I have been giving away to others I’ve been inspired by Autoresearch at night, freedom all day.

I can’t believe someone just gave me 4 RTX pro 6000s, it’s absolutely one of the greatest acts of generosity someone has shown me. The money I would have spent on the cards I have been giving away to others I’ve been inspired by Autoresearch at night, freedom all day.

65,996 görüntüleme • 20 gün önce

GLM-5.2 Nvidia NVFP4 No pruning, no quantising, exact same model card. 110 tok/s single stream. My friend is a genius, holy fudge. Here's the repo, he got full DS4-Flash on a 5090 + DDR5 at 38.5 tok/s This is a breakthrough.

GLM-5.2 Nvidia NVFP4 No pruning, no quantising, exact same model card. 110 tok/s single stream. My friend is a genius, holy fudge. Here's the repo, he got full DS4-Flash on a 5090 + DDR5 at 38.5 tok/s This is a breakthrough.

157,562 görüntüleme • 1 ay önce

Pi agent deep dive, this is a live session. Enjoy! I think there's a lot of good info in there

Pi agent deep dive, this is a live session. Enjoy! I think there's a lot of good info in there

32,592 görüntüleme • 12 gün önce

Cloud on the left Local on the right Amazing times, look at that speed.

Cloud on the left Local on the right Amazing times, look at that speed.

26,564 görüntüleme • 12 gün önce

This is a real task I gave to Qwen3.6-35B running on my DGX Spark, take 3 videos stitch them, and speed the first by 4x. This is something i have been using LLMs for over the last few months, Qwen was flawless! It contributed to local studio fixing the alignment issue u see

This is a real task I gave to Qwen3.6-35B running on my DGX Spark, take 3 videos stitch them, and speed the first by 4x. This is something i have been using LLMs for over the last few months, Qwen was flawless! It contributed to local studio fixing the alignment issue u see

88,997 görüntüleme • 2 ay önce

I can’t believe something like this could come out of a model with no vision GLM-5.3 rocks

I can’t believe something like this could come out of a model with no vision GLM-5.3 rocks

25,941 görüntüleme • 17 gün önce

Kimi IS COOKING holy mackerel this is way better than anything I can get out of opus or GPT Has some bugs.. but looks soooo unique and well into my brand, for 1 shot I can't complain.

235,591 görüntüleme • 7 ay önce

Religion bricks GPT. I'm getting 5 tokens a second, I only have 3 active sessions. This never happened anywhere else. Interesting pattern.

Religion bricks GPT. I'm getting 5 tokens a second, I only have 3 active sessions. This never happened anywhere else. Interesting pattern.

25,775 görüntüleme • 19 gün önce

I promise this is the last pleometric video I drop on you.

I promise this is the last pleometric video I drop on you.

131,800 görüntüleme • 5 ay önce

Preparing to de-Apple my life OG Homelab doing work

Preparing to de-Apple my life OG Homelab doing work

10,309 görüntüleme • 9 gün önce

Holy moly, MiniMax-M2.7 is amazing, watch till the end.

Holy moly, MiniMax-M2.7 is amazing, watch till the end.

110,351 görüntüleme • 5 ay önce

If you're a frontend dev let me make your life easier: 1. Pay for mobbin. 2. Setup mobbin MCP 3. Model will be able to pull all designs off the screens 4. Basically tons of free Figma designs

If you're a frontend dev let me make your life easier: 1. Pay for mobbin. 2. Setup mobbin MCP 3. Model will be able to pull all designs off the screens 4. Basically tons of free Figma designs

74,149 görüntüleme • 3 ay önce

GLM-5.2 on 4x 6000s in ZCode /goal build a nintendo ds emulator in C++, do not use a ready made emulator. Don't stop until you can run the Mario 64 DS rom.

GLM-5.2 on 4x 6000s in ZCode /goal build a nintendo ds emulator in C++, do not use a ready made emulator. Don't stop until you can run the Mario 64 DS rom.

39,351 görüntüleme • 1 ay önce

Here’s everything I’ve learned about Local AI in the last year or so. Grateful to Presidio Bitcoin

Here’s everything I’ve learned about Local AI in the last year or so. Grateful to Presidio Bitcoin

33,852 görüntüleme • 1 ay önce

Local AI can be so good, but you’d need about 12k USD to get it. Then it’s not so great Here’s a Q4 of Qwen3.5-262B-REAP Weights 131GB KVCache 50GB 256,000 context 350 tokens/s prefill warmup 4,000 tokens/s prefill cache 36 tokens/s generation Vision enabled REAP is good

Local AI can be so good, but you’d need about 12k USD to get it. Then it’s not so great Here’s a Q4 of Qwen3.5-262B-REAP Weights 131GB KVCache 50GB 256,000 context 350 tokens/s prefill warmup 4,000 tokens/s prefill cache 36 tokens/s generation Vision enabled REAP is good

87,959 görüntüleme • 5 ay önce

An autist in its natural habitat

An autist in its natural habitat

13,038 görüntüleme • 16 gün önce