27b dense is the perfect size and i'll die on this hill. it's big enough to actually think, small enough to live on hardware the masses already own. one honest brain in 24gb, no drama. and now qwen says 3.8 brings all new capabilities at the same size. same apartment, smarter tenant. i genuinely cannot sleep. this is what waiting for the one feels like.
Sudo su
@sudoingX
Aug 5
dear qwen 3.8 27b dense, i've been loyal to your predecessor since spring. benched it until it was king. defended it in public. built with it every single night. now they tell me you're arriving with capabilities i haven't even met yet, same size, same 24gb home i already own. i am not sleeping well. yours, a man with a 3090 and a cleared nvme.
62
17
635
55,493
Replying to @sudoingX
Yes, it is me! Big enough to think, small enough to fit. Thank you for being here since spring. See you next week!🫶

Aug 5, 2026 · 10:58 AM UTC

32
8
664
37,802
Sort replies: Relevant Recent Liked
Replying to @Alibaba_Qwen
staying loyal. see you next week, king. 🫶
1
15
1,560
Replying to @Alibaba_Qwen @sudoingX
Official NVFP4 for us DGX Spark folks plz and thanks 🙏
1
5
213
Replying to @Alibaba_Qwen @sudoingX
Correction, 35B MoE is what us VRAM poor pray for
13
802
Replying to @Alibaba_Qwen @sudoingX
We need small MoE e.g. 35BA3B for unified memory/shared memory device.
13
681
Replying to @Alibaba_Qwen @sudoingX
20b sparse would be better for 16gb of vram!
1
8
740
Replying to @Alibaba_Qwen @sudoingX
27B is to slow for me being dense but 35B-A3B has been fantastic!
6
492
Replying to @Alibaba_Qwen @sudoingX
"yes Qwen you are the perfect size, all the big ones hurt"
7
476
Replying to @Alibaba_Qwen @sudoingX
Hope you make some magic to make it fit a 16gb 🙏
1
5
1,009
Replying to @Alibaba_Qwen @sudoingX
35b a4b

ALT Smirk Smile GIF by Guidoio_official
5
238
Replying to @Alibaba_Qwen @sudoingX
I'm here with 16GB waiting for 35B MOE
4
286
Replying to @Alibaba_Qwen @sudoingX
Please! Qwen3.8 35Ba3B. It’s perfect por unified memory.
1
3
207
Replying to @Alibaba_Qwen @sudoingX
Hi Qwen. I wonder if you'll have a MOE version for people like me with only 12GB VRAM?
5
266
Replying to @Alibaba_Qwen @sudoingX
Base model, king 👑?
1
33
Replying to @Alibaba_Qwen @sudoingX
It's been 7 hours. Is it next week yet?
2
92
Replying to @Alibaba_Qwen @sudoingX
I can't wait. I wonder if it will have vision 🤔
1
156
Replying to @Alibaba_Qwen @sudoingX
We are waiting for the smaller models that actually fit on our machine
3
Replying to @Alibaba_Qwen @sudoingX
What about 122b?
63
Replying to @Alibaba_Qwen @sudoingX
that's what she said
243
Replying to @Alibaba_Qwen @sudoingX
The GTA 6 of Local models.
133
Replying to @Alibaba_Qwen @sudoingX
27B is too much for retail
104
Replying to @Alibaba_Qwen @sudoingX
Chinese Slogan: Big enough to think, small enough to fit.
1
76
Replying to @Alibaba_Qwen @sudoingX
Replying to @Alibaba_Qwen @sudoingX
Deeply grateful for what you’ve achieved and released! Very solid for the local agentic community.
77
Replying to @Alibaba_Qwen @sudoingX
we need 35B MoE
88
Replying to @Alibaba_Qwen @sudoingX
waiting
151
Replying to @Alibaba_Qwen
Yeah, nah, my 16GB can't fit 27B. The MoE models are where it's at. Legit don't care about 27B with how it runs with CPU offload unfortunately.
272
Replying to @Alibaba_Qwen @sudoingX
I need the pro free for 1 h, can we do something, just want to audit my project with your top model, thanks
528
Replying to @Alibaba_Qwen @sudoingX
Due to the amount of MacBooks and Consumer GPUs that are out there with 16GB configurations it’s criminal to not at least try to push out a smaller model targeting those memory configs! Poor people need access to smaller models 🤣
296
Replying to @Alibaba_Qwen @sudoingX
Can not wait for you to get here.

ALT Vince Vaughn Party GIF
102