GeForce RTX 5070

[...] ...plus the recent price increases by AI companies, made me actually think the opposite: that there might be another additional "run" for memory and/or GPUs. Therefore, yesterday I decided to order an additional RTX 5060 with 16 GiB VRAM for the 500$ that I saved during the last months (to be added to the RTX 5070 12 GiB that I bought last year to play games in 4k + my old RTX 3060 12 GiB which I recycled a few months ago after noticing how nice it is to run llama.cpp locally without having to worry about subscription costs). [...]

[...] It’s built as it’s own backend and philosophy wise, I want my personal AI to connect to everything in my life and have complete context over all data that I own, be compiled into a neat stack of data accumulating, then when it thinks it’s appropriate to do/say something, it’ll do it without my involvement. Nice card btw, I got my RTX 5070 a couple months ago and it runs like a dream.

[...] I'm currently running those models using an RTX 5070 12GiB + RTX 5060 16GiB + RTX 3060 12GiB with a 96k context size with MTP/speculative decoding and I'm quite happy (the 5070 is about 4x faster than the 3060, the 5060 is inbetween them so about 2x faster than a 3060).

I'm using 4x RTX 5070's and first-gen AMD threadripper (1950X) to run Qwen3.6 27B (MTP) Q6K with llama.cpp and it works great as a daily driver with Pi. [...] It's slower being a dense model but the quality seems much better. [...] When I fire up Pi, working with the model is very snappy at start. When I interact with the LLM via Hermes CLI, it's much slower. [...]

Nvidia on laptops? Insert the famous Linus Torvalds meme here I have an RTX 5070 (whatever the laptop variant is) and it absolutely rocks with almost everything I throw at it, running Ubuntu+Steam+Proton. I no longer worry whether a Windows game is going to run, because almost all of them do with good performance.

[...] I have a Ryzen 9800X3D with RTX 5070, 128GB of RAM and TBs of Gen 5.0 NVMes. [...]

I got the Costco deal but was very excited to see what she’s made of AMD Ryzen 7 9800X3D processor ASUS GeForce RTX 5070 12gb XPG lancer blade ddr5 32gb RAM MSI pro b650-VC mobo It’s been so long since I’ve had a new pc 😅 but it looks like an alright build

I bought a new graphics card (RTX 5070) for my birthday, but the card was too big for my case. Using a dremel and aviation shears, I mutilated my case and fit this monster of a graphics card.

I got a Beelink GTR 9 Pro for $1980. These Strix Halo systems were a good deal at $2k when the alternative was a DGX Spark (which is similarly memory-constrained, but has about twice the iGPU processing power of the Radeon 8060S, having as many CUDA cores as an RTX 5070) for $4k. The pitch was basically "half the GPU compute (negative), x86 instead of ARM (positive), but no CUDA (negative), for half the price (positive), but you also don't get the ConnectX-7 NIC (negative)". These more or less balanced out to being worth it if you wanted a single-node system that could also double as a generic x86 homelab server once it was obsolete for LLM workloads. These days, you can get a DGX Spark for $4.7k, so yes, the price has risen, but Strix Halo (with a few exceptions like the Bosgame and Corsair systems) $4k (or more!) is simply not a very good deal. [...]

[...] My performance when using an RTX 5070 12GiB VRAM, Ryzen 7 9700X 8 cores CPU, 32GiB DDR5 6000MT (2 sticks): - "qwen2.5:7b": 128 tokens/second (this model fits 100% in the VRAM). [...] - qwen3.5:35b-a3b: 17 tokens/second, but it's highly unstable and crashes - currently not usable for me. [...] I would therefore deduce that the most important thing is the amount of VRAM and that performance would be similar even when using an older GPU (e.g. [...] Performance without a GPU, tested by using a Ryzen 9 5950X 16 cores CPU, 128GiB DDR4 3200 MT: - "qwen2.5:7b": 9 tokens/second - "qwen3:32b": 2 tokens/second - "qwen3:30b-a3b": 16 tokens/second

Having a single big fan cool a massive heatsink (that is hopefully very quiet) can legitimately a good reason to get this over building a typical SFF PC, which often runs hot and loud. It sorta reminds me of the trashcan Mac Pro. I myself have a sandwich style case with an RTX 5070 in it which is quite loud under load.

I got a new PC for editing videos. Ryzen 9950X3D, 128Gb of RAM, multiple NVMe drives, GeForce RTX 5070. [...]

The readme opens with this: I have an RTX 5070 with 12 GB VRAM and I wanted to run glm-4.7-flash:q80, which is a 31.8 GB model. [...]

[...] There's no way an RTX 5070 needs the same cooler as a 5080/5090 by any stretch, but many of these cards on the lower-mid range are using the same coolers as on the high end. [...] But there are plenty of SFF builders that would like a more reasonably sized upper mid-range card. [...]