highperfocused/ollama
1
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2025-11-10 22:28:43 +01:00
Code Issues Packages Projects Releases Wiki Activity

update llama.cpp submodule to 1e6f6554 (#6208)

Browse Source
This commit is contained in:
Jeffrey Morgan
2024-08-06 15:11:45 -04:00
committed by GitHub
parent d4a7216c82
commit e04c7012c2
4 changed files with 25 additions and 45 deletions
Download Patch File Download Diff File Expand all files Collapse all files

14
llm/ext_server/server.cpp vendored
View File

@@ -403,7 +403,9 @@ struct llama_server_context
}
}
std::tie(model, ctx) = llama_init_from_gpt_params(params);
auto init_result = llama_init_from_gpt_params(params);
model = init_result.model;
ctx = init_result.context;
if (model == nullptr)
{
LOG_ERROR("unable to load model", {{"model", params.model}});
@@ -2422,7 +2424,10 @@ static void server_params_parse(int argc, char **argv, server_params &sparams, g
invalid_param = true;
break;
}
params.lora_adapter.emplace_back(argv[i], 1.0f);
params.lora_adapters.push_back({
std::string(argv[i]),
1.0,
});
params.use_mmap = false;
}
else if (arg == "--lora-scaled")
@@ -2438,7 +2443,10 @@ static void server_params_parse(int argc, char **argv, server_params &sparams, g
invalid_param = true;
break;
}
params.lora_adapter.emplace_back(lora_adapter, std::stof(argv[i]));
params.lora_adapters.push_back({
lora_adapter,
std::stof(argv[i])
});
params.use_mmap = false;
}
else if (arg == "-v" || arg == "--verbose")