3 ms·
I just checked gpt-oss:20b on my M4 Pro 24GB, and got 400.67 tokens/s on input and 46.53 tokens/s on output. That's for a tiny context of 72 tokens.
by onion2k 1y ago
I just checked gpt-oss:20b on my M4 Pro 24GB, and got 400.67 tokens/s on input and 46.53 tokens/s on output. That's for a tiny context of 72 tokens.
- sebastiennight 1y agoThis message was amazing and I want about to hit [New Tab] and purchase one myself until the penultimate word.