https://github.com/ggerganov/llama.cpp/pull/11448

Author: Animaxx

src/llama-vocab.h

@@ -34,7 +34,13 @@ struct llama_vocab {
     std::vector<id>    cache_special_tokens;
     std::vector<token> cache_token_to_piece; // llama_token_to_piece(special = true);
 
-    std::map<std::pair<std::string, std::string>, int> bpe_ranks;
+    struct pair_hash {
+         size_t operator()(const std::pair<std::string, std::string> & p) const {
+             return std::hash<std::string>{}(p.first) ^  //create some hash for pair
+                    (std::hash<std::string>{}(p.second) << 1);
+         }
+     };
+     std::unordered_map<std::pair<std::string, std::string>, int, pair_hash> bpe_ranks;
 
     // default LLaMA special tokens
     // TODO: should we set all of these to LLAMA_TOKEN_NULL?

src/llama.cpp

@@ -4960,7 +4960,7 @@ struct llama_model_loader {
             for (const auto & file : files) {
                 auto * reg = ggml_backend_dev_backend_reg(ggml_backend_dev_by_type(GGML_BACKEND_DEVICE_TYPE_CPU));
                 auto * is_numa_fn = (decltype(ggml_is_numa) *) ggml_backend_reg_get_proc_address(reg, "ggml_backend_cpu_is_numa");
-                std::unique_ptr<llama_mmap> mapping(new llama_mmap(file.get(), prefetch ? -1 : 0, is_numa_fn()));
+                std::unique_ptr<llama_mmap> mapping = std::make_unique<llama_mmap>(file.get(), prefetch ? -1 : 0, is_numa_fn());
                 mmaps_used.emplace_back(mapping->size, 0);
                 if (mlock_mmaps) {
                     std::unique_ptr<llama_mlock> mlock_mmap(new llama_mlock());