I see that GPT has been updated, and I’ve got to say that for once, it is doing 98% of what I want it to do, and how I want it to do it. I’m impressed.
In the less-than-impressed arena, LM Studio is proving to be unusable. The fault lies in my hardware, a 16MB MBP — it just can’t handle the large LLMs. It spits out some of what I want, close to how I want it, and then just says, “enough!” and quits.
Open WebUI is better, running on my 32GB linux server, but still not up to serious tasks.
#LLM
modean987 /
npub1zn…9m5j8
2024-09-10 18:36:43
Author Public Key
npub1zn0k8g44xfe5paq8r60wpw5sp29lpgea2kgkfkpj8xmzp5y42xesv9m5j8Published at
2024-09-10 18:36:43Event JSON
{
"id": "bf1dc5c74e25dd7fed99a234da96187e0c9e83758dbb5ae2a56d8d124c4c56f1",
"pubkey": "14df63a2b5327340f4071e9ee0ba900a8bf0a33d559164d83239b620d09551b3",
"created_at": 1725993403,
"kind": 1,
"tags": [
[
"t",
"LLM"
],
[
"proxy",
"https://mastodon.world/users/modean987/statuses/113114703710228349",
"activitypub"
]
],
"content": "I see that GPT has been updated, and I’ve got to say that for once, it is doing 98% of what I want it to do, and how I want it to do it. I’m impressed.\n\nIn the less-than-impressed arena, LM Studio is proving to be unusable. The fault lies in my hardware, a 16MB MBP — it just can’t handle the large LLMs. It spits out some of what I want, close to how I want it, and then just says, “enough!” and quits.\n\nOpen WebUI is better, running on my 32GB linux server, but still not up to serious tasks.\n\n#LLM",
"sig": "35786159b896ad948133af48c62d0d041f4e4fe1503b56fc3091cc1ccc08a11a328b839700e76066bf085e92b1f7a9a7c1448ee2854f5f01d52a96f63fb31d61"
}