hodlbod on Nostr: I'm pretty sure the benchmarks it advertises are manufactured as well. DeepSeek seems ...
I'm pretty sure the benchmarks it advertises are manufactured as well. DeepSeek seems substantially dumber than other LLMs, just from anecdotal evidence.
Published at
2025-01-30 16:19:38Event JSON
{
"id": "84ab488df5066b7d8e9312673a92a909abcfa85c4cbd58df3ccb9dc6db857fbf",
"pubkey": "97c70a44366a6535c145b333f973ea86dfdc2d7a99da618c40c64705ad98e322",
"created_at": 1738253978,
"kind": 1,
"tags": [
[
"e",
"63a17550d15baae34dd4c363d9e9f6ad9792705caddae025f61bee373e1d72cf",
"wss://nos.lol/",
"root",
"97c70a44366a6535c145b333f973ea86dfdc2d7a99da618c40c64705ad98e322"
],
[
"client",
"Coracle",
"31990:97c70a44366a6535c145b333f973ea86dfdc2d7a99da618c40c64705ad98e322:1685968093690"
]
],
"content": "I'm pretty sure the benchmarks it advertises are manufactured as well. DeepSeek seems substantially dumber than other LLMs, just from anecdotal evidence.",
"sig": "432527614ef2d858c1492d044c618bdb0a8a73dc3a2e730e8e5ebbf673af5733cf9374df8d5a53944f0a1be4197d1c37780db6b9347cc691ed00af2882c6f7f2"
}
