quinta :ubuntu: on Nostr: DeepSeek-V3 now runs at 20 tokens per second on Mac Studio, and that’s a nightmare ...
Published at
2025-03-27 21:28:48Event JSON
{
"id": "d19c0d453850ab950df648f52fe8ed176ddf91ed6362317aa58ad1f107a87aeb",
"pubkey": "05d9f1c2fa7af0e9d8360669b056148449669fc3f9739c206da917ed6feee053",
"created_at": 1743110928,
"kind": 1,
"tags": [
[
"proxy",
"https://mastodon.uno/users/quinta/statuses/114236517785894239",
"activitypub"
]
],
"content": "DeepSeek-V3 now runs at 20 tokens per second on Mac Studio, and that’s a nightmare for OpenAI https://blog.quintarelli.it/2025/03/deepseek-v3-now-runs-at-20-tokens-per-second-on-mac-studio-and-thats-a-nightmare-for-openai/",
"sig": "705e88f5c47932548f0842daf5152a76e8821723f12f0c609fbb661ffdeea3dd1424c04ced69e482681aadbe450e3e33d223bcdd5cde73ed0a814a32989a284d"
}
