Tune Laguna sampling defaults

README.md

@@ -277,10 +277,10 @@ to ordinary decoding when speculation is not paying, so leaving `--dflash`
 enabled is safe.
 
 The shipped GGUF is configured for a 262144-token context. Laguna defaults to
-temperature 1.0, top-k 20, top-p 1.0, and min-p 0; explicit sampling options
-always take precedence. Use `--nothink` or the `laguna-s-2.1-chat` server alias
-for direct replies, and preserve reasoning content between tool calls when
-building a client.
+temperature 0.7, top-k 20, top-p 0.95, and min-p 0.05; explicit sampling
+options always take precedence. Use `--nothink` or the
+`laguna-s-2.1-chat` server alias for direct replies, and preserve reasoning
+content between tool calls when building a client.
 
 Then build:
 

ds4.c

@@ -60543,8 +60543,10 @@ void ds4_engine_sampling_defaults(ds4_engine *e, float *temperature,
         *top_p = 0.95f;
         *min_p = 0.0f;
     } else if (DS4_MODEL_FAMILY == DS4_MODEL_FAMILY_LAGUNA) {
+        *temperature = 0.7f;
         *top_k = 20;
-        *min_p = 0.0f;
+        *top_p = 0.95f;
+        *min_p = 0.05f;
     }
 }
 

ds4_help.c

@@ -207,7 +207,7 @@ static void print_sampling(FILE *fp, const help_colors *c, bool full) {
     opt(fp, c, "--top-p F", "Nucleus sampling probability.");
     opt(fp, c, "--min-p F", "Keep tokens scoring at least F times the top token.");
     opt(fp, c, "--seed N", "Sampling seed for reproducible non-greedy runs.");
-    para(fp, c, "GLM defaults to temperature 1.0, top-p 0.95, and min-p 0. Laguna defaults to temperature 1.0, top-k 20, top-p 1.0, and min-p 0. Explicit options always win.");
+    para(fp, c, "GLM defaults to temperature 1.0, top-p 0.95, and min-p 0. Laguna defaults to temperature 0.7, top-k 20, top-p 0.95, and min-p 0.05. Explicit options always win.");
     opt(fp, c, "--think", "Use normal thinking mode.");
     opt(fp, c, "--think-max", "Use Think Max when context is large enough.");
     opt(fp, c, "--nothink", "Disable thinking and ask for direct replies.");