GeForce RTX 5070 Ti

i feel m6 is good for daily tasks and its price is okay compared to m5. where m5 is best for local models because of its unified memory of 64GB but its expensive compared to NVIDIA Geforce RTX 5070Ti. with 64 GB you can fit models where it won't even load in 16GB VRAM.

Google Earth is a fantastic computer program (which is being retired June 2027; stupid IMHO); Alphabet, Inc., wants you to instead "use the in-browser version of earth.google.com for a better experience." When I attempt to load earth.google.com it is total dogshit, just a terrible all-around experience (with glitches and an additional layer of obfuscation between map and browsing: a 3rd-party web browser). My RTX5070Ti renders everything so beautifully within the Google Earth program – smooth and fully-3D everything. The only reasoning I can muster is that it's harder to embed advertisements/tracking within the program (compared to the website interface); but when I cannot use the application anymore, I will no longer use Google Earth (entirely) – which is unfortunate because it is a fantastic resource (I have a computer dedicated almost-entirely to Google Earth tasks; it is the only machine in my office that can contact Google servers).

[...] Specs for this are AMD 9800X3D, 32GB of system RAM, RTX 5070Ti 16GB

[...] On Cityscapes (TensorRT FP16, RTX 5070 Ti) it comes out ahead of YOLO26 and RF-DETR on detection/instance-segmentation F1 and takes the top mIoU on semantic segmentation, at real-time latency with 2–3x fewer params. [...] Full protocol and every known asymmetry is public here: It's not a fork - the detection core follows the D-FINE paper, but the segmentation heads, training, export and inference are implemented from scratch. [...]

My understanding is that the precision is supposed to help with noise. Less turbulence, etc. FWIW, in my setup (10th gen i5, RTX 5070 Ti in an old Define R3 case), the 12 cm Noctua G2 fans run quieter and have a much less obnoxious noise than the old P/F series, which wipe the floor with the Arctic fan I bought for a computer that lives in the basement and sounds like it's about to take off.

[...] I have an RTX 5070TI laptop. 95% I use it with Tumbleweed. [...]

Interesting that SparseLoCo held up at 72B scale with permissionless participants. I run distributed inference across multiple machines over Tailscale (M2 Max + RTX 5070 Ti), and even in that controlled setup, network variance is the dominant bottleneck. The fact that they got competitive quality with peers joining and leaving freely on 1.1T tokens is impressive — though I'd love to see how much the blockchain verification overhead actually cost in effective compute utilization.

I got the GeForce RTX 5070 Ti ! I won't lie I don't know much about parts, my friend put it together, I just know how to build it lmao but I hope that means smth to you :D

Exactly; when I first got my RTX 5070 Ti (16gb, to game with!!!, upgrading from VEGA56), I loaded then-latest Qwen3.6 (30B, cannot remember exactly). [...]

[...] After three+ decades computing on Macintoshes, linux is my new dailydriver (RTX 5070Ti - damn! [...]

I can run Qwen3.6-35B-A3B at 20 TPS on my laptop with RTX 5070 Ti, with partial offloading to RAM. [...] For mundane tasks like extracting and formatting specific structured text, I use Gemini in Google search

I held my nose and bought an RTX 5070 Ti for $100 over MSRP in January. [...]

[...] I put its performance at roughly equivalent to the RTX 5070 Ti. The M3 Ultra 512 GB for me is about half the performance of the RTX 5070 Ti but obviously it has the ability to do more because of the increased memory. [...]