High-Performance Computing
Workstations & Servers Show off your awesome HPC or Puget station! Use this sub to discuss commercial-grade stuff, higher-end servers, or get help for related topics. Machine Learning, LLMs, & AI A place to post results / benchmarks, best setups, guides, result analysis, and performance questions. E.g. OpenAI, Gemini Multiphysics & Simulation A place to post compiling time results / benchmarks, best setups, guides, result analysis, and performance questions. E.g. COMSOL, SOLIDWORKS
| Topic | Replies | Views | Activity | |
|---|---|---|---|---|
| Upgrade from RX 9700 XT to AI PRO R9700, is it worth it? What are the compromises? |
|
14 | 407 | November 8, 2026 |
| Splitting up work to be split across mutiple slots on the same gpu for higher total tokens per second |
|
12 | 167 | November 12, 2026 |
| The Demon Hidden Inside Qwen (I Pushed One Neuron) |
|
6 | 179 | November 7, 2026 |
| LLM under Proxmox with GPU passthrough? |
|
6 | 218 | November 11, 2026 |
| Strix Halo (EVO-X2 96GB) + Threadripper 3-GPU (7900 XTX / 6750 XT / RX 5700) local LLM homelab: benchmarks, and is retrying eGPUs on Strix Halo worth it? |
|
6 | 208 | November 8, 2026 |
| Zml/llmd: a new fast LLM server in Zig |
|
0 | 62 | November 7, 2026 |
| Crowd-sourcing Model Training: Anyone have any experience with CrowdGPT? |
|
0 | 45 | November 8, 2026 |