@minisforum.official: We tested the MINISFORUM UM790 Pro with Qwen 35B — and the results might surprise you. 💻 Powered by: • AMD Ryzen™ 9 7940HS • Radeon™ 780M iGPU • 48GB DDR5 RAM (dual-channel) 🚀 Smooth local inference at 20+ tokens/sec (based on our test setup, results may vary) No cloud. No API fees. Just local AI. 🔒 #MINISFORUM #UM790Pro #LocalAI #MiniPC #AI  

Minisforum
Minisforum
Open In TikTok:
Region: US
Friday 24 April 2026 06:29:53 GMT
30864
513
38
120

Music

Download

Comments

truckyand
Longavid :
$350 plus $350 for 48gb ram
2026-04-25 02:06:39
13
dquestion.alsoal
D'question Al-So'al :
bullshit, the ram has to be vram, not normal ram, this could be a scam
2026-04-25 12:55:26
1
wakandanomore123456
Wak Anda For Eva :
i stop using ollama when my laptop only produce 8 tokens a second
2026-05-02 01:17:00
2
frankynouva_ios
FrankyNouva_iOS :
Only > 80 tk/s are worth buying 🤣
2026-04-25 05:09:16
2
thangtraniot
thang tran :
tóm lại cần cái máy nào share ram cho card màn hình nhiều thi se dùng đc model lớn? 48gb ram share cho igpu 32 Mb là đc 32Vram? hay bản chất chạy llama.ccp này thì nó đã load 8Gb mô hình ko dùng vào ram?
2026-04-24 17:23:58
0
daniel_dslr
𝕯𝖆𝖓𝖎𝖊𝖑 :
Can test Gemma 4?
2026-04-24 21:19:41
0
kokrramolles
An avocado :
That RAM is slow as fuck for an LLM
2026-04-27 20:49:51
0
hellmy492
Joes :
need 5090🗿
2026-04-24 12:12:12
0
carwash2016
CarWash2016 :
Rubbish $900 with ram and hard disk at least
2026-04-26 21:09:04
0
enwo1911
Enrico Wolf39 :
Difference Gemma 4 can save use cache while qwen and llama is crash bug and needs to rebuild always the complete prompt
2026-04-26 09:19:21
0
y1ngc7
穎熙@y1 :
the problem now is ram too gad dang expensive
2026-05-13 06:51:08
0
mr.miguelm
user2864865273889 :
W/h ??? How much does it consume in energy?
2026-04-24 21:27:54
0
drachenvodzen
WarDaMalHaarDran :
22 tokens 😂 Okay when you can wait for minutes on answers
2026-04-25 15:35:57
0
akuveer
veer 勇毅 อาวเขียว :
Can you test on Fedora as well?
2026-04-30 13:08:37
1
youngpapi.com
youngpapi :
can I get result for qwen2.5-coder qwen3-coder and qwen3
2026-05-06 12:21:10
0
bobbadung
bobnapibadung :
thank you for the demo include with the settings. what is uma frame buffer for?
2026-04-25 13:07:55
0
nguyen.mike3
Nguyen Mike :
just invested 1 vga 16G Vram
2026-05-01 10:03:03
0
sattanax
Sattanax :
$18k mxn
2026-04-27 04:58:28
0
_k8s_j
_k8s_j :
it does work with Linux?
2026-05-02 03:53:36
0
nomad_nam
The Nẫm :
wtf, i got 64gb ram. this is new. let's test.
2026-04-25 11:57:04
0
youngpapi.com
youngpapi :
price
2026-05-06 12:19:20
0
antonio.c547
Eduardo Antonio Cruz Rodríguez :
😂😂😂
2026-05-14 01:05:08
0
To see more videos from user @minisforum.official, please go to the Tikwm homepage.

Other Videos


About