mirror of
https://github.com/ggml-org/llama.cpp.git
synced 2026-07-15 00:45:56 +02:00
97af49fa39
* server : reuse llama_sample_token common function * common : use n_probs for temperature sampling