LaunchDevelopers
24 days ago
llama.cpp b9297 adds NVFP4 and MTP support
llama.cpp version b9297 adds support for NVFP4 quantization and Multi-Token Prediction (MTP). The update is available now on GitHub.
·
24 days ago
llama.cpp version b9297 adds support for NVFP4 quantization and Multi-Token Prediction (MTP). The update is available now on GitHub.