llama.cpp\src\llama-kv-cache-unified.cpp:226: GGML_ASSERT(seq_id >= 0 && (size_t) seq_id < seq_to_stream.size()) failed

#4
by devold - opened

llama.cpp\src\llama-kv-cache-unified.cpp:226: GGML_ASSERT(seq_id >= 0 && (size_t) seq_id < seq_to_stream.size()) failed

Consider reporting it to llama.cpp. The latest release works for me.

llama.cpp\src\llama-kv-cache-unified.cpp:226: GGML_ASSERT(seq_id >= 0 && (size_t) seq_id < seq_to_stream.size()) failed

Redownload and rebuild

Sign up or log in to comment