RWKV / rwkv.cpp

INT4/INT5/INT8 and FP16 inference on CPU for RWKV language model
MIT License
1.41k stars 95 forks source link

Update new GGML for GGML_MAX_NODES limit? #142

Open fann1993814 opened 11 months ago

fann1993814 commented 11 months ago

I see this. https://github.com/ggerganov/ggml/issues/567

The new GGML has removed GGML_MAX_NODES limit. It is more friendly for RNN-based model seemly.

saharNooby commented 11 months ago

Hi! Thanks for the heads up. I update upstream ggml once in a couple of months, so this change will inevitably be integrated at some point.

fann1993814 commented 11 months ago

Recently rwkv-v5 has been released. Maybe update the network flow to adapt the new ggml framework, There must be a lot of works to be done. 🧐