3 ms·

Isn’t llama.cpp’s implementation of Qwen 3.5 better, or am I misinformed?

• antirez 7mo ago
  There was a recent fix by ollama and I used it.