bump llama.cpp to b7356 (#501)

Author: ilopezluna

llamacpp/native/src/server/server-common.cpp

@@ -494,6 +494,18 @@ int32_t server_tokens::process_chunk(
     return 0;
 }
 
+server_tokens server_tokens::clone() const {
+    server_tokens res;
+    res.has_mtmd = has_mtmd;
+    res.tokens   = tokens;
+    for (auto it = map_idx_to_media.begin(); it != map_idx_to_media.end(); ++it) {
+        size_t idx = it->first;
+        const mtmd::input_chunk_ptr & chunk = it->second;
+        res.map_idx_to_media[idx] = mtmd::input_chunk_ptr(mtmd_input_chunk_copy(chunk.get()));
+    }
+    return res;
+}
+
 //
 // tokenizer and input processing utils
 //
@@ -745,12 +757,6 @@ json oaicompat_completion_params_parse(const json & body) {
         llama_params["stop"] = json_value(body, "stop", json::array());
     }
 
-    // Handle "n" field
-    int n_choices = json_value(body, "n", 1);
-    if (n_choices != 1) {
-        throw std::runtime_error("Only one completion choice is allowed");
-    }
-
     // Handle "echo" field
     if (json_value(body, "echo", false)) {
         throw std::runtime_error("Only no echo is supported");
@@ -791,7 +797,7 @@ static void handle_media(
         SRV_INF("downloading image from '%s'\n", url.c_str());
         auto res = common_remote_get_content(url, params);
         if (200 <= res.first && res.first < 300) {
-            SRV_INF("downloaded %ld bytes\n", res.second.size());
+            SRV_INF("downloaded %zu bytes\n", res.second.size());
             raw_buffer data;
             data.insert(data.end(), res.second.begin(), res.second.end());
             out_files.push_back(data);
@@ -966,6 +972,9 @@ json oaicompat_chat_params_parse(
     inputs.parallel_tool_calls   = json_value(body, "parallel_tool_calls", false);
     inputs.add_generation_prompt = json_value(body, "add_generation_prompt", true);
     inputs.reasoning_format      = opt.reasoning_format;
+    if (body.contains("reasoning_format")) {
+        inputs.reasoning_format = common_reasoning_format_from_name(body.at("reasoning_format").get<std::string>());
+    }
     inputs.enable_thinking       = opt.enable_thinking;
     if (!inputs.tools.empty() && inputs.tool_choice != COMMON_CHAT_TOOL_CHOICE_NONE) {
         if (body.contains("grammar")) {
@@ -1045,11 +1054,8 @@ json oaicompat_chat_params_parse(
     for (const auto & stop : chat_params.additional_stops) {
         llama_params["stop"].push_back(stop);
     }
-
-    // Handle "n" field
-    int n_choices = json_value(body, "n", 1);
-    if (n_choices != 1) {
-        throw std::invalid_argument("Only one completion choice is allowed");
+    if (!chat_params.parser.empty()) {
+        llama_params["chat_parser"] = chat_params.parser;
     }
 
     // Handle "logprobs" field

llamacpp/native/src/server/server-common.h

@@ -18,11 +18,13 @@ const static std::string build_info("b" + std::to_string(LLAMA_BUILD_NUMBER) + "
 using json = nlohmann::ordered_json;
 
 #define SLT_INF(slot, fmt, ...) LOG_INF("slot %12.*s: id %2d | task %d | " fmt, 12, __func__, (slot).id, ((slot).task ? (slot).task->id : -1), __VA_ARGS__)
+#define SLT_CNT(slot, fmt, ...) LOG_CNT(""                                 fmt,                                                                __VA_ARGS__)
 #define SLT_WRN(slot, fmt, ...) LOG_WRN("slot %12.*s: id %2d | task %d | " fmt, 12, __func__, (slot).id, ((slot).task ? (slot).task->id : -1), __VA_ARGS__)
 #define SLT_ERR(slot, fmt, ...) LOG_ERR("slot %12.*s: id %2d | task %d | " fmt, 12, __func__, (slot).id, ((slot).task ? (slot).task->id : -1), __VA_ARGS__)
 #define SLT_DBG(slot, fmt, ...) LOG_DBG("slot %12.*s: id %2d | task %d | " fmt, 12, __func__, (slot).id, ((slot).task ? (slot).task->id : -1), __VA_ARGS__)
 
 #define SRV_INF(fmt, ...) LOG_INF("srv  %12.*s: " fmt, 12, __func__, __VA_ARGS__)
+#define SRV_CNT(fmt, ...) LOG_CNT(""              fmt,               __VA_ARGS__)
 #define SRV_WRN(fmt, ...) LOG_WRN("srv  %12.*s: " fmt, 12, __func__, __VA_ARGS__)
 #define SRV_ERR(fmt, ...) LOG_ERR("srv  %12.*s: " fmt, 12, __func__, __VA_ARGS__)
 #define SRV_DBG(fmt, ...) LOG_DBG("srv  %12.*s: " fmt, 12, __func__, __VA_ARGS__)
@@ -215,6 +217,8 @@ struct server_tokens {
                 llama_pos pos,
                 int32_t seq_id,
                 size_t & n_tokens_out) const;
+
+    server_tokens clone() const;
 };