nikola/llama_cpp
1
0
Fork 0
Code Issues Pull Requests Actions 18 Packages Projects Releases Wiki Activity

read n_ctx back after making llama_context (#21939)

Browse Source
This commit is contained in:
Matt
2026-04-15 15:24:57 +08:00
committed by GitHub
parent 5d14e5d19b
commit e39eba26f3
1 changed files with 2 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files
examples/diffusion/diffusion-cli.cpp
+2 -2
View File
@@ -602,8 +602,8 @@ int main(int argc, char ** argv) {
int n_input = input_tokens.size(); int n_input = input_tokens.size();
if (n_input >= params.n_ctx) { if (static_cast<uint32_t>(n_input) >= llama_n_ctx(ctx)) {
LOG_ERR("error: input too long (%d tokens), max context is %d\n", n_input, params.n_ctx); LOG_ERR("error: input too long (%d tokens), max context is %d\n", n_input, llama_n_ctx(ctx));
llama_free(ctx); llama_free(ctx);
llama_model_free(model); llama_model_free(model);
return 1; return 1;