Track_Shovel@slrpnk.net to Lemmy Shitpost@lemmy.worldEnglish · 5 days agoHexadecimalslrpnk.netimagemessage-square142fedilinkarrow-up11.06Karrow-down125
arrow-up11.03Karrow-down1imageHexadecimalslrpnk.netTrack_Shovel@slrpnk.net to Lemmy Shitpost@lemmy.worldEnglish · 5 days agomessage-square142fedilink
minus-squarecoldsideofyourpillow@lemmy.cafelinkfedilinkEnglisharrow-up1·2 days agoBy running it locally. The local models don’t have any censorship.
minus-squarevvilld@lemmy.worldlinkfedilinkarrow-up1·2 days agoI meant, how does one run it locally. I see a lot of people saying to just “run it locally” but for someone without a background in coding that doesn’t really mean much.
minus-squarecoldsideofyourpillow@lemmy.cafelinkfedilinkEnglisharrow-up1·edit-22 days agoYou don’t need a background in coding at all. In fact, the spaces of machine learning and programming are almost completely seperate. Download Ollama. Depending on the power of your GPU, run one of the following commands: DeepSeek-R1-Distill-Qwen-1.5B: ollama run deepseek-r1:1.5b DeepSeek-R1-Distill-Qwen-7B: ollama run deepseek-r1:7b DeepSeek-R1-Distill-Llama-8B: ollama run deepseek-r1:8b DeepSeek-R1-Distill-Qwen-14B: ollama run deepseek-r1:14b DeepSeek-R1-Distill-Qwen-32B: ollama run deepseek-r1:32b DeepSeek-R1-Distill-Llama-70B: ollama run deepseek-r1:70b Bigger models means better output, but also longer generation times.
minus-squareCharlxmagne@lemmy.worldlinkfedilinkarrow-up1·2 days agoThey do by default but like I said it’s open source so you can tweak it to not be.
By running it locally. The local models don’t have any censorship.
I meant, how does one run it locally. I see a lot of people saying to just “run it locally” but for someone without a background in coding that doesn’t really mean much.
You don’t need a background in coding at all. In fact, the spaces of machine learning and programming are almost completely seperate.
Download Ollama.
Depending on the power of your GPU, run one of the following commands:
DeepSeek-R1-Distill-Qwen-1.5B:
ollama run deepseek-r1:1.5bDeepSeek-R1-Distill-Qwen-7B:
ollama run deepseek-r1:7bDeepSeek-R1-Distill-Llama-8B:
ollama run deepseek-r1:8bDeepSeek-R1-Distill-Qwen-14B:
ollama run deepseek-r1:14bDeepSeek-R1-Distill-Qwen-32B:
ollama run deepseek-r1:32bDeepSeek-R1-Distill-Llama-70B:
ollama run deepseek-r1:70bBigger models means better output, but also longer generation times.
They do by default but like I said it’s open source so you can tweak it to not be.