6 ms·

fastllm targets the GPU, while colibri uses CPU inference only

• aliljet 2mo ago
  I'd be curious about an.option that would allow glm use with a low end GPU like a 2080 ti...