server: accept null sampling params (#25538)

* server: accept null sampling params

Extend the schema validation to treat a null value as absent, so
clients can send null on nullable params (temperature, top_p, ...)
to request the server default. This matches the OpenAI spec and the
json_value convention used elsewhere.

Add has_field() to skip null in the field eval guards.

* has_field -> has_value​

src/llama-model.cpp

@@ -336,38 +336,38 @@ struct ggml_backend_meta_split_state llama_meta_device_get_split_state(const str
     const llama_hparams & hparams = ud->model->hparams;
     const std::string tensor_name = tensor->name;
 
-    const std::regex pattern_q_weight        ("blk\\.\\d*\\.attn_q.weight");
-    const std::regex pattern_kv_weight       ("blk\\.\\d*\\.attn_(k|v).weight");
-    const std::regex pattern_qkv_weight      ("blk\\.\\d*\\.attn_qkv.weight");
-    const std::regex pattern_q_bias          ("blk\\.\\d*\\.attn_q\\.bias");
-    const std::regex pattern_kv_bias         ("blk\\.\\d*\\.attn_(k|v)\\.bias");
-    const std::regex pattern_qkv_bias        ("blk\\.\\d*\\.attn_qkv.bias");
-    const std::regex pattern_qk_norm         ("blk\\.\\d*\\.attn_(q|k)_norm\\.weight");
-    const std::regex pattern_kv_cache        ("cache_(k|v)_l\\d*");
-    const std::regex pattern_attn_sinks      ("blk\\.\\d*\\.attn_sinks.weight");
-    const std::regex pattern_attn_out_weight ("blk\\.\\d*\\.attn_output.weight");
-    const std::regex pattern_attn_out_bias   ("blk\\.\\d*\\.attn_output.bias");
-    const std::regex pattern_attn_gate_weight("blk\\.\\d*\\.attn_gate.weight");
+    static const std::regex pattern_q_weight        ("blk\\.\\d*\\.attn_q.weight");
+    static const std::regex pattern_kv_weight       ("blk\\.\\d*\\.attn_(k|v).weight");
+    static const std::regex pattern_qkv_weight      ("blk\\.\\d*\\.attn_qkv.weight");
+    static const std::regex pattern_q_bias          ("blk\\.\\d*\\.attn_q\\.bias");
+    static const std::regex pattern_kv_bias         ("blk\\.\\d*\\.attn_(k|v)\\.bias");
+    static const std::regex pattern_qkv_bias        ("blk\\.\\d*\\.attn_qkv.bias");
+    static const std::regex pattern_qk_norm         ("blk\\.\\d*\\.attn_(q|k)_norm\\.weight");
+    static const std::regex pattern_kv_cache        ("cache_(k|v)_l\\d*");
+    static const std::regex pattern_attn_sinks      ("blk\\.\\d*\\.attn_sinks.weight");
+    static const std::regex pattern_attn_out_weight ("blk\\.\\d*\\.attn_output.weight");
+    static const std::regex pattern_attn_out_bias   ("blk\\.\\d*\\.attn_output.bias");
+    static const std::regex pattern_attn_gate_weight("blk\\.\\d*\\.attn_gate.weight");
 
-    const std::regex pattern_ssm_dt          ("blk\\.\\d*\\.ssm_dt.bias");
-    const std::regex pattern_ssm_a           ("blk\\.\\d*\\.ssm_a");
-    const std::regex pattern_ssm_alpha       ("blk\\.\\d*\\.ssm_alpha.weight");
-    const std::regex pattern_ssm_beta        ("blk\\.\\d*\\.ssm_beta.weight");
-    const std::regex pattern_ssm_beta_alpha  ("blk\\.\\d*\\.ssm_ba.weight");
-    const std::regex pattern_r_cache         ("cache_r_l\\d*");
-    const std::regex pattern_s_cache         ("cache_s_l\\d*");
-    const std::regex pattern_ssm_conv1d      ("blk\\.\\d*\\.ssm_conv1d.weight");
-    const std::regex pattern_ssm_out_weight  ("blk\\.\\d*\\.ssm_out.weight");
+    static const std::regex pattern_ssm_dt          ("blk\\.\\d*\\.ssm_dt.bias");
+    static const std::regex pattern_ssm_a           ("blk\\.\\d*\\.ssm_a");
+    static const std::regex pattern_ssm_alpha       ("blk\\.\\d*\\.ssm_alpha.weight");
+    static const std::regex pattern_ssm_beta        ("blk\\.\\d*\\.ssm_beta.weight");
+    static const std::regex pattern_ssm_beta_alpha  ("blk\\.\\d*\\.ssm_ba.weight");
+    static const std::regex pattern_r_cache         ("cache_r_l\\d*");
+    static const std::regex pattern_s_cache         ("cache_s_l\\d*");
+    static const std::regex pattern_ssm_conv1d      ("blk\\.\\d*\\.ssm_conv1d.weight");
+    static const std::regex pattern_ssm_out_weight  ("blk\\.\\d*\\.ssm_out.weight");
 
-    const std::regex pattern_ffn_up_gate_weight("blk\\.\\d*\\.ffn_(up|gate)(_exps)?.weight");
-    const std::regex pattern_ffn_up_gate_bias  ("blk\\.\\d*\\.ffn_(up|gate)(_exps)?.bias");
-    const std::regex pattern_ffn_gate_up_weight("blk\\.\\d*\\.ffn_gate_up(_exps)?.weight");
-    const std::regex pattern_ffn_down_weight   ("blk\\.\\d*\\.ffn_down(_exps)?.weight");
-    const std::regex pattern_ffn_down_bias     ("blk\\.\\d*\\.ffn_down.bias");
-    const std::regex pattern_ffn_down_exps_bias("blk\\.\\d*\\.ffn_down_exps.bias");
+    static const std::regex pattern_ffn_up_gate_weight("blk\\.\\d*\\.ffn_(up|gate)(_exps)?.weight");
+    static const std::regex pattern_ffn_up_gate_bias  ("blk\\.\\d*\\.ffn_(up|gate)(_exps)?.bias");
+    static const std::regex pattern_ffn_gate_up_weight("blk\\.\\d*\\.ffn_gate_up(_exps)?.weight");
+    static const std::regex pattern_ffn_down_weight   ("blk\\.\\d*\\.ffn_down(_exps)?.weight");
+    static const std::regex pattern_ffn_down_bias     ("blk\\.\\d*\\.ffn_down.bias");
+    static const std::regex pattern_ffn_down_exps_bias("blk\\.\\d*\\.ffn_down_exps.bias");
 
-    const std::regex pattern_output_weight("output\\.weight");
-    const std::regex pattern_output_bias  ("output\\.bias");
+    static const std::regex pattern_output_weight("output\\.weight");
+    static const std::regex pattern_output_bias  ("output\\.bias");
 
     struct tensor_config {
         ggml_backend_meta_split_axis axis;

tools/server/server-schema.cpp

@@ -568,10 +568,16 @@ static void handle_with_catch(const char * name, std::function<void()> func) {
     }
 }
 
+// treat a null value as absent so clients can send null to request the server default
+static bool has_value(const json & data, const char * n) {
+    auto it = data.find(n);
+    return it != data.end() && !it->is_null();
+}
+
 template <typename T>
 void field_num<T>::eval(field_eval_context & ctx, const json & data) {
     for (const auto & n : name) {
-        if (data.contains(n)) {
+        if (has_value(data, n)) {
             handle_with_catch(n, [&]() {
                 if (custom_handler) {
                 custom_handler(ctx, data);
@@ -593,7 +599,7 @@ void field_num<T>::eval(field_eval_context & ctx, const json & data) {
 void field_str::eval(field_eval_context & ctx, const json & data) {
     GGML_ASSERT(custom_handler);
     for (const auto & n : name) {
-        if (data.contains(n)) {
+        if (has_value(data, n)) {
             handle_with_catch(n, [&]() {
                 custom_handler(ctx, data);
             });
@@ -604,7 +610,7 @@ void field_str::eval(field_eval_context & ctx, const json & data) {
 
 void field_bool::eval(field_eval_context & ctx, const json & data) {
     for (const auto & n : name) {
-        if (data.contains(n)) {
+        if (has_value(data, n)) {
             handle_with_catch(n, [&]() {
                 if (custom_handler) {
                     custom_handler(ctx, data);
@@ -620,7 +626,7 @@ void field_bool::eval(field_eval_context & ctx, const json & data) {
 void field_json::eval(field_eval_context & ctx, const json & data) {
     GGML_ASSERT(custom_handler);
     for (const auto & n : name) {
-        if (data.contains(n)) {
+        if (has_value(data, n)) {
             handle_with_catch(n, [&]() {
                 custom_handler(ctx, data);
             });