Merge branch 'upstream' into concedo_experimental

# Conflicts:
#	.devops/full-cuda.Dockerfile
#	.devops/full-rocm.Dockerfile
#	.devops/llama-cli-cuda.Dockerfile
#	.devops/llama-cli-rocm.Dockerfile
#	.devops/llama-cli-vulkan.Dockerfile
#	.devops/llama-cpp-cuda.srpm.spec
#	.devops/llama-server-cuda.Dockerfile
#	.devops/llama-server-rocm.Dockerfile
#	.devops/llama-server-vulkan.Dockerfile
#	.github/workflows/build.yml
#	.github/workflows/docker.yml
#	CMakeLists.txt
#	Makefile
#	README.md
#	examples/llama.android/llama/src/main/cpp/CMakeLists.txt
#	flake.lock
#	ggml/CMakeLists.txt
#	ggml/src/CMakeLists.txt
#	grammars/README.md
#	scripts/sync-ggml-am.sh
#	scripts/sync-ggml.last
#	tests/test-chat-template.cpp
#	tests/test-grammar-integration.cpp
#	tests/test-json-schema-to-grammar.cpp

examples/infill/README.md

@@ -15,6 +15,7 @@ In this section, we cover the most commonly used options for running the `infill
 -   `-i, --interactive`: Run the program in interactive mode, allowing you to provide input directly and receive real-time responses.
 -   `-n N, --n-predict N`: Set the number of tokens to predict when generating text. Adjusting this value can influence the length of the generated text.
 -   `-c N, --ctx-size N`: Set the size of the prompt context. The default is 512, but LLaMA models were built with a context of 2048, which will provide better results for longer input/inference.
+-   `--spm-infill`: Use Suffix/Prefix/Middle pattern for infill (instead of Prefix/Suffix/Middle) as some models prefer this.
 
 ## Input Prompts
 

examples/infill/infill.cpp

@@ -211,6 +211,7 @@ int main(int argc, char ** argv) {
         suff_rm_leading_spc = false;
     }
     std::vector<llama_token> embd_inp;
+    std::vector<llama_token> embd_end;
     std::vector<llama_token> inp_pfx = ::llama_tokenize(ctx, params.input_prefix, false);
     std::vector<llama_token> inp_sfx = ::llama_tokenize(ctx, params.input_suffix, false);
     const int space_token = 29871;
@@ -218,12 +219,13 @@ int main(int argc, char ** argv) {
         inp_sfx.erase(inp_sfx.begin());
     }
     inp_pfx.insert(inp_pfx.begin(), llama_token_prefix(model));
-    if (add_bos) {
-        inp_pfx.insert(inp_pfx.begin(), llama_token_bos(model));
-    }
     inp_sfx.insert(inp_sfx.begin(), llama_token_suffix(model));
-    embd_inp = inp_pfx;
-    embd_inp.insert(embd_inp.end(), inp_sfx.begin(), inp_sfx.end());
+    embd_inp = params.spm_infill ? inp_sfx : inp_pfx;
+    embd_end = params.spm_infill ? inp_pfx : inp_sfx;
+    if (add_bos) {
+        embd_inp.insert(embd_inp.begin(), llama_token_bos(model));
+    }
+    embd_inp.insert(embd_inp.end(), embd_end.begin(), embd_end.end());
 
     const llama_token middle_token = llama_token_middle(model);
     if (middle_token >= 0) {
@@ -527,14 +529,14 @@ int main(int argc, char ** argv) {
                     inp_sfx.erase(inp_sfx.begin());
                 }
                 inp_pfx.insert(inp_pfx.begin(), llama_token_prefix(model));
-                if (add_bos) {
-                    inp_pfx.insert(inp_pfx.begin(), llama_token_bos(model));
-                }
                 inp_sfx.insert(inp_sfx.begin(), llama_token_suffix(model));
-                embd_inp = inp_pfx;
-                embd_inp.insert(embd_inp.end(), inp_sfx.begin(), inp_sfx.end());
+                embd_inp = params.spm_infill ? inp_sfx : inp_pfx;
+                embd_end = params.spm_infill ? inp_pfx : inp_sfx;
+                if (add_bos) {
+                    embd_inp.insert(embd_inp.begin(), llama_token_bos(model));
+                }
+                embd_inp.insert(embd_inp.end(), embd_end.begin(), embd_end.end());
 
-                const llama_token middle_token = llama_token_middle(model);
                 if (middle_token >= 0) {
                     embd_inp.push_back(middle_token);
                 }