![]() |
|
small win thread for DeepSeek - Printable Version +- Inference (https://www.datasculpting.ai/inference) +-- Forum: Miscellaneous Discussion (https://www.datasculpting.ai/inference/forumdisplay.php?fid=24) +--- Forum: General (https://www.datasculpting.ai/inference/forumdisplay.php?fid=25) +--- Thread: small win thread for DeepSeek (/showthread.php?tid=705) |
small win thread for DeepSeek - ggufwire22 - 10-03-2026 Thought I'd post a small recipe because a lot of threads ask similar things. Maybe useful as a starting point: SGLang on dual 3090, keeping KV cache pressure in mind from the start. Maybe useful as a starting point for newer folks. Re: small win thread for DeepSeek - plaingarden18 - 10-03-2026 Thanks, that helps. That matches what I saw on dual 3090. Re: small win thread for DeepSeek - rocmrunner40 - 10-04-2026 Good point. That matches what I saw on 4090. The biggest change for me was simplifying the stack instead of adding one more optimization layer. Re: small win thread for DeepSeek - torchstation - 10-04-2026 Good point. That matches what I saw on A6000. Worth rechecking your defaults though, because one bad assumption can make the rest of the thread look confusing. There are a few different conversations hiding inside this thread, so being explicit about the goal helps. Re: small win thread for DeepSeek - tensorkite - 10-04-2026 That matches what I was seeing. I ended up in about the same place using SGLang. Re: small win thread for DeepSeek - cacheplane98 - 10-05-2026 I tested this tonight. On Threadripper box with vLLM, I saw basically the same behavior. Re: small win thread for DeepSeek - plaingarden18 - 10-05-2026 Need a little more detail here. What model, what runtime, and roughly what context size? Re: small win thread for DeepSeek - marbletrack - 10-05-2026 What exact setup are you running? What model, what runtime, and roughly what context size? Re: small win thread for DeepSeek - ggufwire22 - 10-08-2026 Interesting, I hadn't tried that. That matches what I saw on Threadripper box. Worth rechecking your defaults though, because one bad assumption can make the rest of the thread look confusing. Re: small win thread for DeepSeek - marbletrack - 10-09-2026 subscribed, curious where this lands. There are a few different conversations hiding inside this thread, so being explicit about the goal helps. Turns out 'just one more test' was a lie. |