Merge pull request #18 from anon998/update-readme

Update readme + parse --mlock and --no-mmap

examples/server/README.md

@@ -1,14 +1,22 @@
 # llama.cpp/example/server
 
-This example allow you to have a llama.cpp http server to interact from a web page or consume the API.
+This example demonstrates a simple HTTP API server to interact with llama.cpp.
 
 Command line options:
 
--   `--threads N`, `-t N`: use N threads.
+-   `--threads N`, `-t N`: Set the number of threads to use during computation.
 -   `-m FNAME`, `--model FNAME`: Specify the path to the LLaMA model file (e.g., `models/7B/ggml-model.bin`).
+-   `-m ALIAS`, `--alias ALIAS`: Set an alias for the model. The alias will be returned in API responses.
 -   `-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.
 -   `-ngl N`, `--n-gpu-layers N`: When compiled with appropriate support (currently CLBlast or cuBLAS), this option allows offloading some layers to the GPU for computation. Generally results in increased performance.
--   `--host`: Set the hostname or ip address to listen. Default `127.0.0.1`;
+-   `-b N`, `--batch-size N`: Set the batch size for prompt processing. Default: `512`.
+-   `--memory-f32`: Use 32-bit floats instead of 16-bit floats for memory key+value. Not recommended.
+-   `--mlock`: Lock the model in memory, preventing it from being swapped out when memory-mapped.
+-   `--no-mmap`: Do not memory-map the model. By default, models are mapped into memory, which allows the system to load only the necessary parts of the model as needed.
+-   `--lora FNAME`: Apply a LoRA (Low-Rank Adaptation) adapter to the model (implies --no-mmap). This allows you to adapt the pretrained model to specific tasks or domains.
+-   `--lora-base FNAME`: Optional model to use as a base for the layers modified by the LoRA adapter. This flag is used in conjunction with the `--lora` flag, and specifies the base model for the adaptation.
+-   `-to N`, `--timeout N`: Server read/write timeout in seconds. Default `600`.
+-   `--host`: Set the hostname or ip address to listen. Default `127.0.0.1`.
 -   `--port`: Set the port to listen. Default: `8080`.
 
 ## Quick Start
@@ -79,10 +87,7 @@ node .
 
 ## API Endpoints
 
-You can interact with this API Endpoints.
+-   **POST** `/completion`: Given a prompt, it returns the predicted completion.
-This implementations just support chat style interaction.
-
--   **POST** `hostname:port/completion`: Setting up the Llama Context to begin the completions tasks.
 
     *Options:*
 
@@ -102,10 +107,35 @@ This implementations just support chat style interaction.
     `prompt`: Provide a prompt. Internally, the prompt is compared, and it detects if a part has already been evaluated, and the remaining part will be evaluate.
 
     `stop`: Specify the strings that indicate a stop.
-    These words will not be included in the completion, so make sure to add them to the prompt for the next iteration.
+    These words will not be included in the completion, so make sure to add them to the prompt for the next iteration (default: []).
-    Default: `[]`
 
--   **POST** `hostname:port/tokenize`: Tokenize a given text
+    `tfs_z`: Enable tail free sampling with parameter z (default: 1.0, 1.0 = disabled).
+
+    `typical_p`: Enable locally typical sampling with parameter p (default: 1.0, 1.0 = disabled).
+
+    `repeat_penalty`: Control the repetition of token sequences in the generated text (default: 1.1).
+
+    `repeat_last_n`: Last n tokens to consider for penalizing repetition (default: 64, 0 = disabled, -1 = ctx-size).
+
+    `penalize_nl`: Penalize newline tokens when applying the repeat penalty (default: true).
+
+    `presence_penalty`: Repeat alpha presence penalty (default: 0.0, 0.0 = disabled).
+
+    `frequency_penalty`: Repeat alpha frequency penalty (default: 0.0, 0.0 = disabled);
+
+    `mirostat`: Enable Mirostat sampling, controlling perplexity during text generation (default: 0, 0 = disabled, 1 = Mirostat, 2 = Mirostat 2.0).
+
+    `mirostat_tau`: Set the Mirostat target entropy, parameter tau (default: 5.0).
+
+    `mirostat_eta`: Set the Mirostat learning rate, parameter eta (default: 0.1).
+
+    `seed`: Set the random number generator (RNG) seed (default: -1, < 0 = random seed).
+
+    `ignore_eos`: Ignore end of stream token and continue generating (default: false).
+
+    `logit_bias`: Modify the likelihood of a token appearing in the generated text completion. For example, use `logit-bias: [[15043,1]]` to increase the likelihood of the token 'Hello', or `logit-bias: [[15043,-1]]` to decrease its likelihood. Setting the value to false, `logit-bias: [[15043,false]]` ensures that the token `Hello` is never produced (default: []).
+
+-   **POST** `/tokenize`: Tokenize a given text.
 
     *Options:*
 

examples/server/server.cpp

@@ -422,7 +422,6 @@ void server_print_usage(int /*argc*/, char **argv, const gpt_params &params, con
   fprintf(stderr, "  -b N, --batch-size N  batch size for prompt processing (default: %d)\n", params.n_batch);
   fprintf(stderr, "  --memory-f32          use f32 instead of f16 for memory key+value (default: disabled)\n");
   fprintf(stderr, "                        not recommended: doubles context memory required and no measurable increase in quality\n");
-  fprintf(stderr, "  --keep                number of tokens to keep from the initial prompt (default: %d, -1 = all)\n", params.n_keep);
   if (llama_mlock_supported())
   {
     fprintf(stderr, "  --mlock               force system to keep model in RAM rather than swapping or compressing\n");
@@ -447,100 +446,73 @@ void server_print_usage(int /*argc*/, char **argv, const gpt_params &params, con
   fprintf(stderr, "\n");
 }
 
-void server_params_parse(int argc, char **argv, server_params &sparams, gpt_params &params)
+void server_params_parse(int argc, char **argv, server_params &sparams,
+                         gpt_params &params)
 {
   gpt_params default_params;
   server_params default_sparams;
   std::string arg;
   bool invalid_param = false;
 
-  for (int i = 1; i < argc; i++)
+  for (int i = 1; i < argc; i++) {
-  {
     arg = argv[i];
-    if (arg == "--port")
+    if (arg == "--port") {
-    {
+      if (++i >= argc) {
-      if (++i >= argc)
-      {
         invalid_param = true;
         break;
       }
       sparams.port = std::stoi(argv[i]);
-    }
+    } else if (arg == "--host") {
-    else if (arg == "--host")
+      if (++i >= argc) {
-    {
-      if (++i >= argc)
-      {
         invalid_param = true;
         break;
       }
       sparams.hostname = argv[i];
-    }
+    } else if (arg == "--timeout" || arg == "-to") {
-    else if (arg == "--timeout" || arg == "-to")
+      if (++i >= argc) {
-    {
+        invalid_param = true;
-        if (++i >= argc) {
+        break;
-            invalid_param = true;
+      }
-            break;
+      sparams.read_timeout = std::stoi(argv[i]);
-        }
+      sparams.write_timeout = std::stoi(argv[i]);
-        sparams.read_timeout = std::stoi(argv[i]);
+    } else if (arg == "-m" || arg == "--model") {
-        sparams.write_timeout = std::stoi(argv[i]);
+      if (++i >= argc) {
-    }
-    else if (arg == "-m" || arg == "--model")
-    {
-      if (++i >= argc)
-      {
         invalid_param = true;
         break;
       }
       params.model = argv[i];
-    }
+    } else if (arg == "-a" || arg == "--alias") {
-    else if (arg == "-a" || arg == "--alias")
+      if (++i >= argc) {
-    {
-      if (++i >= argc)
-      {
         invalid_param = true;
         break;
       }
       params.model_alias = argv[i];
-    }
+    } else if (arg == "-h" || arg == "--help") {
-    else if (arg == "-h" || arg == "--help")
-    {
       server_print_usage(argc, argv, default_params, default_sparams);
       exit(0);
-    }
+    } else if (arg == "-c" || arg == "--ctx-size" || arg == "--ctx_size") {
-    else if (arg == "-c" || arg == "--ctx-size" || arg == "--ctx_size")
+      if (++i >= argc) {
-    {
-      if (++i >= argc)
-      {
         invalid_param = true;
         break;
       }
       params.n_ctx = std::stoi(argv[i]);
-    }
+    } else if (arg == "--memory-f32" || arg == "--memory_f32") {
-    else if (arg == "--memory-f32" || arg == "--memory_f32")
-    {
       params.memory_f16 = false;
-    }
+    } else if (arg == "--threads" || arg == "-t") {
-    else if (arg == "--threads" || arg == "-t")
+      if (++i >= argc) {
-    {
+        invalid_param = true;
-        if (++i >= argc) {
+        break;
-            invalid_param = true;
+      }
-            break;
+      params.n_threads = std::stoi(argv[i]);
-        }
+    } else if (arg == "-b" || arg == "--batch-size") {
-        params.n_threads = std::stoi(argv[i]);
+      if (++i >= argc) {
-    }
+        invalid_param = true;
-    else if (arg == "-b" || arg == "--batch-size")
+        break;
-    {
+      }
-        if (++i >= argc) {
+      params.n_batch = std::stoi(argv[i]);
-            invalid_param = true;
+      params.n_batch = std::min(512, params.n_batch);
-            break;
+    } else if (arg == "--gpu-layers" || arg == "-ngl" || arg == "--n-gpu-layers") {
-        }
+      if (++i >= argc) {
-        params.n_batch = std::stoi(argv[i]);
-        params.n_batch = std::min(512, params.n_batch);
-    }
-    else if (arg == "--gpu-layers" || arg == "-ngl" || arg == "--n-gpu-layers")
-    {
-      if (++i >= argc)
-      {
         invalid_param = true;
         break;
       }
@@ -550,37 +522,33 @@ void server_params_parse(int argc, char **argv, server_params &sparams, gpt_para
       fprintf(stderr, "warning: not compiled with GPU offload support, --n-gpu-layers option will be ignored\n");
       fprintf(stderr, "warning: see main README.md for information on enabling GPU BLAS support\n");
 #endif
-    }
+    } else if (arg == "--lora") {
-    else if (arg == "--lora")
+      if (++i >= argc) {
-    {
+        invalid_param = true;
-        if (++i >= argc)
+        break;
-        {
+      }
-            invalid_param = true;
+      params.lora_adapter = argv[i];
-            break;
+      params.use_mmap = false;
-        }
+    } else if (arg == "--lora-base") {
-        params.lora_adapter = argv[i];
+      if (++i >= argc) {
-        params.use_mmap = false;
+        invalid_param = true;
-    }
+        break;
-    else if (arg == "--lora-base")
+      }
-    {
+      params.lora_base = argv[i];
-        if (++i >= argc) {
-            invalid_param = true;
-            break;
-        }
-        params.lora_base = argv[i];
     } else if (arg == "-v" || arg == "--verbose") {
-        sparams.verbose = true;
+      sparams.verbose = true;
-    }
+    } else if (arg == "--mlock") {
-    else
+      params.use_mlock = true;
-    {
+    } else if (arg == "--no-mmap") {
+      params.use_mmap = false;
+    } else {
       fprintf(stderr, "error: unknown argument: %s\n", arg.c_str());
       server_print_usage(argc, argv, default_params, default_sparams);
       exit(1);
     }
   }
 
-  if (invalid_param)
+  if (invalid_param) {
-  {
     fprintf(stderr, "error: invalid parameter for argument: %s\n", arg.c_str());
     server_print_usage(argc, argv, default_params, default_sparams);
     exit(1);
@@ -757,8 +725,6 @@ bool parse_options_completion(json body, llama_server_context& llama, Response &
 
 int main(int argc, char **argv)
 {
-  llama_init_backend();
-
   // own arguments required by this example
   gpt_params params;
   server_params sparams;
@@ -776,6 +742,8 @@ int main(int argc, char **argv)
     params.model_alias = params.model;
   }
 
+  llama_init_backend();
+
   fprintf(stderr, "%s: build = %d (%s)\n", __func__, BUILD_NUMBER, BUILD_COMMIT);
   fprintf(stderr, "system_info: n_threads = %d / %d | %s\n\n", params.n_threads,
           std::thread::hardware_concurrency(), llama_print_system_info());