Event JSON
{
"id": "9798caf358f76347bc67b2f92c229b90377a0987ef1e35055bffd676b302893d",
"pubkey": "53f0a7a0eebb00b18e94e17a6a8178d1c346cdd60f0448c409a23ea2c033f8c7",
"created_at": 1754212416,
"kind": 1,
"tags": [
[
"t",
"llamacpp"
],
[
"t",
"ollama"
],
[
"t",
"llama"
],
[
"t",
"llm"
],
[
"t",
"ai"
],
[
"proxy",
"https://m.hirad.it/@redstar/114964064899056211",
"web"
],
[
"proxy",
"https://m.hirad.it/users/redstar/statuses/114964064899056211",
"activitypub"
],
[
"L",
"pink.momostr"
],
[
"l",
"pink.momostr.activitypub:https://m.hirad.it/users/redstar/statuses/114964064899056211",
"pink.momostr"
],
[
"-"
]
],
"content": "I'm considering to test llama.cpp and see if I prefer it over Ollama. \nThe good thing about Ollama is the ease of use, but it's terrible when it comes to model format. Meanwhile llama.cpp even though more complex, it simply works with gguf files, which makes things much easier. \nAlso in terms of performance, llama.cpp is supposed to be better than Ollama. But I have to try it to know for sure. \n\n#Ollama #llamacpp #llama #LLM #AI",
"sig": "47b896c9fa1661f56eac8e95e207d186bc7321c5229053069eb019ed2683a9025c50b7747bee15be3c4e547afd2b9858ce56550a58d36e858ca1cf0bef7398c1"
}