@fugood/llama.node 0.6.2 → 1.0.0-beta.1

package/src/llama.cpp/examples/deprecation-warning/deprecation-warning.cpp

@@ -1,35 +0,0 @@
-// Warns users that this filename was deprecated, and provides a link for more information.
-
-#include <cstdio>
-#include <string>
-#include <unordered_map>
-
-// Main
-int main(int argc, char** argv) {
-    std::string filename = "main";
-    if (argc >= 1) {
-        filename = argv[0];
-    }
-
-    // Get only the program name from the full path
-    auto pos = filename.find_last_of("/\\");
-    if (pos != std::string::npos) {
-        filename = filename.substr(pos+1);
-    }
-
-    // Append "llama-" to the beginning of filename to get the replacemnt filename
-    auto replacement_filename = "llama-" + filename;
-
-    // The exception is if the filename is "main", then our replacement filename is "llama-cli"
-    if (filename == "main") {
-        replacement_filename = "llama-cli";
-    }
-
-    fprintf(stdout, "\n");
-    fprintf(stdout, "WARNING: The binary '%s' is deprecated.\n", filename.c_str());
-    fprintf(stdout, " Please use '%s' instead.\n", replacement_filename.c_str());
-    fprintf(stdout, " See https://github.com/ggerganov/llama.cpp/tree/master/examples/deprecation-warning/README.md for more information.\n");
-    fprintf(stdout, "\n");
-
-    return EXIT_FAILURE;
-}

package/src/llama.cpp/examples/embedding/CMakeLists.txt

@@ -1,5 +0,0 @@
-set(TARGET llama-embedding)
-add_executable(${TARGET} embedding.cpp)
-install(TARGETS ${TARGET} RUNTIME)
-target_link_libraries(${TARGET} PRIVATE common llama ${CMAKE_THREAD_LIBS_INIT})
-target_compile_features(${TARGET} PRIVATE cxx_std_17)

package/src/llama.cpp/examples/embedding/embedding.cpp

@@ -1,323 +0,0 @@
-#include "arg.h"
-#include "common.h"
-#include "log.h"
-#include "llama.h"
-
-#include <ctime>
-#include <algorithm>
-
-#if defined(_MSC_VER)
-#pragma warning(disable: 4244 4267) // possible loss of data
-#endif
-
-static std::vector<std::string> split_lines(const std::string & s, const std::string & separator = "\n") {
-    std::vector<std::string> lines;
-    size_t start = 0;
-    size_t end = s.find(separator);
-
-    while (end != std::string::npos) {
-        lines.push_back(s.substr(start, end - start));
-        start = end + separator.length();
-        end = s.find(separator, start);
-    }
-
-    lines.push_back(s.substr(start)); // Add the last part
-
-    return lines;
-}
-
-static void batch_add_seq(llama_batch & batch, const std::vector<int32_t> & tokens, llama_seq_id seq_id) {
-    size_t n_tokens = tokens.size();
-    for (size_t i = 0; i < n_tokens; i++) {
-        common_batch_add(batch, tokens[i], i, { seq_id }, true);
-    }
-}
-
-static void batch_decode(llama_context * ctx, llama_batch & batch, float * output, int n_seq, int n_embd, int embd_norm) {
-    const enum llama_pooling_type pooling_type = llama_pooling_type(ctx);
-
-    // clear previous kv_cache values (irrelevant for embeddings)
-    llama_kv_self_clear(ctx);
-
-    // run model
-    LOG_INF("%s: n_tokens = %d, n_seq = %d\n", __func__, batch.n_tokens, n_seq);
-    if (llama_encode(ctx, batch) < 0) {
-        LOG_ERR("%s : failed to encode\n", __func__);
-    }
-
-    for (int i = 0; i < batch.n_tokens; i++) {
-        if (!batch.logits[i]) {
-            continue;
-        }
-
-        const float * embd = nullptr;
-        int embd_pos = 0;
-
-        if (pooling_type == LLAMA_POOLING_TYPE_NONE) {
-            // try to get token embeddings
-            embd = llama_get_embeddings_ith(ctx, i);
-            embd_pos = i;
-            GGML_ASSERT(embd != NULL && "failed to get token embeddings");
-        } else {
-            // try to get sequence embeddings - supported only when pooling_type is not NONE
-            embd = llama_get_embeddings_seq(ctx, batch.seq_id[i][0]);
-            embd_pos = batch.seq_id[i][0];
-            GGML_ASSERT(embd != NULL && "failed to get sequence embeddings");
-        }
-
-        float * out = output + embd_pos * n_embd;
-        common_embd_normalize(embd, out, n_embd, embd_norm);
-    }
-}
-
-int main(int argc, char ** argv) {
-    common_params params;
-
-    if (!common_params_parse(argc, argv, params, LLAMA_EXAMPLE_EMBEDDING)) {
-        return 1;
-    }
-
-    common_init();
-
-    params.embedding = true;
-
-    // utilize the full context
-    if (params.n_batch < params.n_ctx) {
-        LOG_WRN("%s: setting batch size to %d\n", __func__, params.n_ctx);
-        params.n_batch = params.n_ctx;
-    }
-
-    // For non-causal models, batch size must be equal to ubatch size
-    params.n_ubatch = params.n_batch;
-
-    llama_backend_init();
-    llama_numa_init(params.numa);
-
-    // load the model
-    common_init_result llama_init = common_init_from_params(params);
-
-    llama_model * model = llama_init.model.get();
-    llama_context * ctx = llama_init.context.get();
-
-    if (model == NULL) {
-        LOG_ERR("%s: unable to load model\n", __func__);
-        return 1;
-    }
-
-    const llama_vocab * vocab = llama_model_get_vocab(model);
-
-    const int n_ctx_train = llama_model_n_ctx_train(model);
-    const int n_ctx = llama_n_ctx(ctx);
-
-    const enum llama_pooling_type pooling_type = llama_pooling_type(ctx);
-
-    if (llama_model_has_encoder(model) && llama_model_has_decoder(model)) {
-        LOG_ERR("%s: computing embeddings in encoder-decoder models is not supported\n", __func__);
-        return 1;
-    }
-
-    if (n_ctx > n_ctx_train) {
-        LOG_WRN("%s: warning: model was trained on only %d context tokens (%d specified)\n",
-                __func__, n_ctx_train, n_ctx);
-    }
-
-    // print system information
-    {
-        LOG_INF("\n");
-        LOG_INF("%s\n", common_params_get_system_info(params).c_str());
-    }
-
-    // split the prompt into lines
-    std::vector<std::string> prompts = split_lines(params.prompt, params.embd_sep);
-
-    // max batch size
-    const uint64_t n_batch = params.n_batch;
-
-    // tokenize the prompts and trim
-    std::vector<std::vector<int32_t>> inputs;
-    for (const auto & prompt : prompts) {
-        auto inp = common_tokenize(ctx, prompt, true, true);
-        if (inp.size() > n_batch) {
-            LOG_ERR("%s: number of tokens in input line (%lld) exceeds batch size (%lld), increase batch size and re-run\n",
-                    __func__, (long long int) inp.size(), (long long int) n_batch);
-            return 1;
-        }
-        inputs.push_back(inp);
-    }
-
-    // check if the last token is SEP
-    // it should be automatically added by the tokenizer when 'tokenizer.ggml.add_eos_token' is set to 'true'
-    for (auto & inp : inputs) {
-        if (inp.empty() || inp.back() != llama_vocab_sep(vocab)) {
-            LOG_WRN("%s: last token in the prompt is not SEP\n", __func__);
-            LOG_WRN("%s: 'tokenizer.ggml.add_eos_token' should be set to 'true' in the GGUF header\n", __func__);
-        }
-    }
-
-    // tokenization stats
-    if (params.verbose_prompt) {
-        for (int i = 0; i < (int) inputs.size(); i++) {
-            LOG_INF("%s: prompt %d: '%s'\n", __func__, i, prompts[i].c_str());
-            LOG_INF("%s: number of tokens in prompt = %zu\n", __func__, inputs[i].size());
-            for (int j = 0; j < (int) inputs[i].size(); j++) {
-                LOG("%6d -> '%s'\n", inputs[i][j], common_token_to_piece(ctx, inputs[i][j]).c_str());
-            }
-            LOG("\n\n");
-        }
-    }
-
-    // initialize batch
-    const int n_prompts = prompts.size();
-    struct llama_batch batch = llama_batch_init(n_batch, 0, 1);
-
-    // count number of embeddings
-    int n_embd_count = 0;
-    if (pooling_type == LLAMA_POOLING_TYPE_NONE) {
-        for (int k = 0; k < n_prompts; k++) {
-            n_embd_count += inputs[k].size();
-        }
-    } else {
-        n_embd_count = n_prompts;
-    }
-
-    // allocate output
-    const int n_embd = llama_model_n_embd(model);
-    std::vector<float> embeddings(n_embd_count * n_embd, 0);
-    float * emb = embeddings.data();
-
-    // break into batches
-    int e = 0; // number of embeddings already stored
-    int s = 0; // number of prompts in current batch
-    for (int k = 0; k < n_prompts; k++) {
-        // clamp to n_batch tokens
-        auto & inp = inputs[k];
-
-        const uint64_t n_toks = inp.size();
-
-        // encode if at capacity
-        if (batch.n_tokens + n_toks > n_batch) {
-            float * out = emb + e * n_embd;
-            batch_decode(ctx, batch, out, s, n_embd, params.embd_normalize);
-            e += pooling_type == LLAMA_POOLING_TYPE_NONE ? batch.n_tokens : s;
-            s = 0;
-            common_batch_clear(batch);
-        }
-
-        // add to batch
-        batch_add_seq(batch, inp, s);
-        s += 1;
-    }
-
-    // final batch
-    float * out = emb + e * n_embd;
-    batch_decode(ctx, batch, out, s, n_embd, params.embd_normalize);
-
-    if (params.embd_out.empty()) {
-        LOG("\n");
-
-        if (pooling_type == LLAMA_POOLING_TYPE_NONE) {
-            for (int j = 0; j < n_embd_count; j++) {
-                LOG("embedding %d: ", j);
-                for (int i = 0; i < std::min(3, n_embd); i++) {
-                    if (params.embd_normalize == 0) {
-                        LOG("%6.0f ", emb[j * n_embd + i]);
-                    } else {
-                        LOG("%9.6f ", emb[j * n_embd + i]);
-                    }
-                }
-                LOG(" ... ");
-                for (int i = n_embd - 3; i < n_embd; i++) {
-                    if (params.embd_normalize == 0) {
-                        LOG("%6.0f ", emb[j * n_embd + i]);
-                    } else {
-                        LOG("%9.6f ", emb[j * n_embd + i]);
-                    }
-                }
-                LOG("\n");
-            }
-        } else if (pooling_type == LLAMA_POOLING_TYPE_RANK) {
-            for (int j = 0; j < n_embd_count; j++) {
-                // NOTE: if you change this log - update the tests in ci/run.sh
-                LOG("rerank score %d: %8.3f\n", j, emb[j * n_embd]);
-            }
-        } else {
-            // print the first part of the embeddings or for a single prompt, the full embedding
-            for (int j = 0; j < n_prompts; j++) {
-                LOG("embedding %d: ", j);
-                for (int i = 0; i < (n_prompts > 1 ? std::min(16, n_embd) : n_embd); i++) {
-                    if (params.embd_normalize == 0) {
-                        LOG("%6.0f ", emb[j * n_embd + i]);
-                    } else {
-                        LOG("%9.6f ", emb[j * n_embd + i]);
-                    }
-                }
-                LOG("\n");
-            }
-
-            // print cosine similarity matrix
-            if (n_prompts > 1) {
-                LOG("\n");
-                LOG("cosine similarity matrix:\n\n");
-                for (int i = 0; i < n_prompts; i++) {
-                    LOG("%6.6s ", prompts[i].c_str());
-                }
-                LOG("\n");
-                for (int i = 0; i < n_prompts; i++) {
-                    for (int j = 0; j < n_prompts; j++) {
-                        float sim = common_embd_similarity_cos(emb + i * n_embd, emb + j * n_embd, n_embd);
-                        LOG("%6.2f ", sim);
-                    }
-                    LOG("%1.10s", prompts[i].c_str());
-                    LOG("\n");
-                }
-            }
-        }
-    }
-
-    if (params.embd_out == "json" || params.embd_out == "json+" || params.embd_out == "array") {
-        const bool notArray = params.embd_out != "array";
-
-        LOG(notArray ? "{\n  \"object\": \"list\",\n  \"data\": [\n" : "[");
-        for (int j = 0;;) { // at least one iteration (one prompt)
-            if (notArray) LOG("    {\n      \"object\": \"embedding\",\n      \"index\": %d,\n      \"embedding\": ",j);
-            LOG("[");
-            for (int i = 0;;) { // at least one iteration (n_embd > 0)
-                LOG(params.embd_normalize == 0 ? "%1.0f" : "%1.7f", emb[j * n_embd + i]);
-                i++;
-                if (i < n_embd) LOG(","); else break;
-            }
-            LOG(notArray ? "]\n    }" : "]");
-            j++;
-            if (j < n_embd_count) LOG(notArray ? ",\n" : ","); else break;
-        }
-        LOG(notArray ? "\n  ]" : "]\n");
-
-        if (params.embd_out == "json+" && n_prompts > 1) {
-            LOG(",\n  \"cosineSimilarity\": [\n");
-            for (int i = 0;;) { // at least two iteration (n_embd_count > 1)
-                LOG("    [");
-                for (int j = 0;;) { // at least two iteration (n_embd_count > 1)
-                    float sim = common_embd_similarity_cos(emb + i * n_embd, emb + j * n_embd, n_embd);
-                    LOG("%6.2f", sim);
-                    j++;
-                    if (j < n_embd_count) LOG(", "); else break;
-                }
-                LOG(" ]");
-                i++;
-                if (i < n_embd_count) LOG(",\n"); else break;
-            }
-            LOG("\n  ]");
-        }
-
-        if (notArray) LOG("\n}\n");
-    }
-
-    LOG("\n");
-    llama_perf_context_print(ctx);
-
-    // clean up
-    llama_batch_free(batch);
-    llama_backend_free();
-
-    return 0;
-}

package/src/llama.cpp/examples/eval-callback/CMakeLists.txt

@@ -1,10 +0,0 @@
-set(TARGET llama-eval-callback)
-add_executable(${TARGET} eval-callback.cpp)
-install(TARGETS ${TARGET} RUNTIME)
-target_link_libraries(${TARGET} PRIVATE common llama ${CMAKE_THREAD_LIBS_INIT})
-target_compile_features(${TARGET} PRIVATE cxx_std_17)
-
-set(TEST_TARGET test-eval-callback)
-add_test(NAME ${TEST_TARGET}
-        COMMAND llama-eval-callback --hf-repo ggml-org/models --hf-file tinyllamas/stories260K.gguf --model stories260K.gguf --prompt hello --seed 42 -ngl 0)
-set_property(TEST ${TEST_TARGET} PROPERTY LABELS eval-callback curl)

package/src/llama.cpp/examples/eval-callback/eval-callback.cpp

@@ -1,194 +0,0 @@
-#include "arg.h"
-#include "common.h"
-#include "log.h"
-#include "llama.h"
-#include "ggml.h"
-
-#include <cstdio>
-#include <string>
-#include <vector>
-
-/**
- * This the arbitrary data which will be passed to each callback.
- * Later on we can for example add operation or tensor name filter from the CLI arg, or a file descriptor to dump the tensor.
- */
-struct callback_data {
-    std::vector<uint8_t> data;
-};
-
-static std::string ggml_ne_string(const ggml_tensor * t) {
-    std::string str;
-    for (int i = 0; i < GGML_MAX_DIMS; ++i) {
-        str += std::to_string(t->ne[i]);
-        if (i + 1 < GGML_MAX_DIMS) {
-            str += ", ";
-        }
-    }
-    return str;
-}
-
-static void ggml_print_tensor(uint8_t * data, ggml_type type, const int64_t * ne, const size_t * nb, int64_t n) {
-    GGML_ASSERT(n > 0);
-    float sum = 0;
-    for (int64_t i3 = 0; i3 < ne[3]; i3++) {
-        LOG("                                     [\n");
-        for (int64_t i2 = 0; i2 < ne[2]; i2++) {
-            if (i2 == n && ne[2] > 2*n) {
-                LOG("                                      ..., \n");
-                i2 = ne[2] - n;
-            }
-            LOG("                                      [\n");
-            for (int64_t i1 = 0; i1 < ne[1]; i1++) {
-                if (i1 == n && ne[1] > 2*n) {
-                    LOG("                                       ..., \n");
-                    i1 = ne[1] - n;
-                }
-                LOG("                                       [");
-                for (int64_t i0 = 0; i0 < ne[0]; i0++) {
-                    if (i0 == n && ne[0] > 2*n) {
-                        LOG("..., ");
-                        i0 = ne[0] - n;
-                    }
-                    size_t i = i3 * nb[3] + i2 * nb[2] + i1 * nb[1] + i0 * nb[0];
-                    float v;
-                    if (type == GGML_TYPE_F16) {
-                        v = ggml_fp16_to_fp32(*(ggml_fp16_t *) &data[i]);
-                    } else if (type == GGML_TYPE_F32) {
-                        v = *(float *) &data[i];
-                    } else if (type == GGML_TYPE_I32) {
-                        v = (float) *(int32_t *) &data[i];
-                    } else if (type == GGML_TYPE_I16) {
-                        v = (float) *(int16_t *) &data[i];
-                    } else if (type == GGML_TYPE_I8) {
-                        v = (float) *(int8_t *) &data[i];
-                    } else {
-                        GGML_ABORT("fatal error");
-                    }
-                    LOG("%12.4f", v);
-                    sum += v;
-                    if (i0 < ne[0] - 1) LOG(", ");
-                }
-                LOG("],\n");
-            }
-            LOG("                                      ],\n");
-        }
-        LOG("                                     ]\n");
-        LOG("                                     sum = %f\n", sum);
-    }
-}
-
-/**
- * GGML operations callback during the graph execution.
- *
- * @param t current tensor
- * @param ask when ask is true, the scheduler wants to know if we are interested in data from this tensor
- *            if we return true, a follow-up call will be made with ask=false in which we can do the actual collection.
- *            see ggml_backend_sched_eval_callback
- * @param user_data user data to pass at each call back
- * @return true to receive data or continue the graph, false otherwise
- */
-static bool ggml_debug(struct ggml_tensor * t, bool ask, void * user_data) {
-    auto * cb_data = (callback_data *) user_data;
-
-    const struct ggml_tensor * src0 = t->src[0];
-    const struct ggml_tensor * src1 = t->src[1];
-
-    if (ask) {
-        return true; // Always retrieve data
-    }
-
-    char src1_str[128] = {0};
-    if (src1) {
-        snprintf(src1_str, sizeof(src1_str), "%s{%s}", src1->name, ggml_ne_string(src1).c_str());
-    }
-
-    LOG("%s: %24s = (%s) %10s(%s{%s}, %s}) = {%s}\n", __func__,
-         t->name, ggml_type_name(t->type), ggml_op_desc(t),
-         src0->name, ggml_ne_string(src0).c_str(),
-         src1 ? src1_str : "",
-         ggml_ne_string(t).c_str());
-
-
-    // copy the data from the GPU memory if needed
-    const bool is_host = ggml_backend_buffer_is_host(t->buffer);
-
-    if (!is_host) {
-        auto n_bytes = ggml_nbytes(t);
-        cb_data->data.resize(n_bytes);
-        ggml_backend_tensor_get(t, cb_data->data.data(), 0, n_bytes);
-    }
-
-    if (!ggml_is_quantized(t->type)) {
-        uint8_t * data = is_host ? (uint8_t *) t->data : cb_data->data.data();
-        ggml_print_tensor(data, t->type, t->ne, t->nb, 3);
-    }
-
-    return true;
-}
-
-static bool run(llama_context * ctx, const common_params & params) {
-    const llama_model * model = llama_get_model(ctx);
-    const llama_vocab * vocab = llama_model_get_vocab(model);
-
-    const bool add_bos = llama_vocab_get_add_bos(vocab);
-
-    std::vector<llama_token> tokens = common_tokenize(ctx, params.prompt, add_bos);
-
-    if (llama_decode(ctx, llama_batch_get_one(tokens.data(), tokens.size()))) {
-        LOG_ERR("%s : failed to eval\n", __func__);
-        return false;
-    }
-
-    return true;
-}
-
-int main(int argc, char ** argv) {
-    callback_data cb_data;
-
-    common_params params;
-
-    if (!common_params_parse(argc, argv, params, LLAMA_EXAMPLE_COMMON)) {
-        return 1;
-    }
-
-    common_init();
-
-    llama_backend_init();
-    llama_numa_init(params.numa);
-
-    // pass the callback to the backend scheduler
-    // it will be executed for each node during the graph computation
-    params.cb_eval = ggml_debug;
-    params.cb_eval_user_data = &cb_data;
-    params.warmup = false;
-
-    // init
-    common_init_result llama_init = common_init_from_params(params);
-
-    llama_model * model = llama_init.model.get();
-    llama_context * ctx = llama_init.context.get();
-
-    if (model == nullptr || ctx == nullptr) {
-        LOG_ERR("%s : failed to init\n", __func__);
-        return 1;
-    }
-
-    // print system information
-    {
-        LOG_INF("\n");
-        LOG_INF("%s\n", common_params_get_system_info(params).c_str());
-        LOG_INF("\n");
-    }
-
-    bool OK = run(ctx, params);
-    if (!OK) {
-        return 1;
-    }
-
-    LOG("\n");
-    llama_perf_context_print(ctx);
-
-    llama_backend_free();
-
-    return 0;
-}

package/src/llama.cpp/examples/gen-docs/CMakeLists.txt

@@ -1,5 +0,0 @@
-set(TARGET llama-gen-docs)
-add_executable(${TARGET} gen-docs.cpp)
-install(TARGETS ${TARGET} RUNTIME)
-target_link_libraries(${TARGET} PRIVATE common llama ${CMAKE_THREAD_LIBS_INIT})
-target_compile_features(${TARGET} PRIVATE cxx_std_17)

package/src/llama.cpp/examples/gen-docs/gen-docs.cpp

@@ -1,83 +0,0 @@
-#include "arg.h"
-#include "common.h"
-
-#include <fstream>
-#include <string>
-
-// Export usage message (-h) to markdown format
-
-static void write_table_header(std::ofstream & file) {
-    file << "| Argument | Explanation |\n";
-    file << "| -------- | ----------- |\n";
-}
-
-static void write_table_entry(std::ofstream & file, const common_arg & opt) {
-    file << "| `";
-    // args
-    for (const auto & arg : opt.args) {
-    if (arg == opt.args.front()) {
-            file << arg;
-            if (opt.args.size() > 1) file << ", ";
-        } else {
-            file << arg << (arg != opt.args.back() ? ", " : "");
-        }
-    }
-    // value hint
-    if (opt.value_hint) {
-        std::string md_value_hint(opt.value_hint);
-        string_replace_all(md_value_hint, "|", "\\|");
-        file << " " << md_value_hint;
-    }
-    if (opt.value_hint_2) {
-        std::string md_value_hint_2(opt.value_hint_2);
-        string_replace_all(md_value_hint_2, "|", "\\|");
-        file << " " << md_value_hint_2;
-    }
-    // help text
-    std::string md_help(opt.help);
-    string_replace_all(md_help, "\n", "<br/>");
-    string_replace_all(md_help, "|", "\\|");
-    file << "` | " << md_help << " |\n";
-}
-
-static void write_table(std::ofstream & file, std::vector<common_arg *> & opts) {
-    write_table_header(file);
-    for (const auto & opt : opts) {
-        write_table_entry(file, *opt);
-    }
-}
-
-static void export_md(std::string fname, llama_example ex) {
-    std::ofstream file(fname, std::ofstream::out | std::ofstream::trunc);
-
-    common_params params;
-    auto ctx_arg = common_params_parser_init(params, ex);
-
-    std::vector<common_arg *> common_options;
-    std::vector<common_arg *> sparam_options;
-    std::vector<common_arg *> specific_options;
-    for (auto & opt : ctx_arg.options) {
-        // in case multiple LLAMA_EXAMPLE_* are set, we prioritize the LLAMA_EXAMPLE_* matching current example
-        if (opt.is_sparam) {
-            sparam_options.push_back(&opt);
-        } else if (opt.in_example(ctx_arg.ex)) {
-            specific_options.push_back(&opt);
-        } else {
-            common_options.push_back(&opt);
-        }
-    }
-
-    file << "**Common params**\n\n";
-    write_table(file, common_options);
-    file << "\n\n**Sampling params**\n\n";
-    write_table(file, sparam_options);
-    file << "\n\n**Example-specific params**\n\n";
-    write_table(file, specific_options);
-}
-
-int main(int, char **) {
-    export_md("autogen-main.md", LLAMA_EXAMPLE_MAIN);
-    export_md("autogen-server.md", LLAMA_EXAMPLE_SERVER);
-
-    return 0;
-}

package/src/llama.cpp/examples/gguf/CMakeLists.txt

@@ -1,5 +0,0 @@
-set(TARGET llama-gguf)
-add_executable(${TARGET} gguf.cpp)
-install(TARGETS ${TARGET} RUNTIME)
-target_link_libraries(${TARGET} PRIVATE ggml ${CMAKE_THREAD_LIBS_INIT})
-target_compile_features(${TARGET} PRIVATE cxx_std_17)