server : tuning tests (#7388)

* server : don't pass temperature as string

* server : increase timeout

* tests : fix the fix 0.8f -> 0.8

ggml-ci

* tests : set explicit temperature

examples/server/tests/features/results.feature

@@ -13,6 +13,7 @@ Feature: Results
 
   Scenario Outline: consistent results with same seed
     Given <n_slots> slots
+    And   0.0 temperature
     Then  the server is starting
     Then  the server is healthy
 
@@ -30,6 +31,7 @@ Feature: Results
 
   Scenario Outline: different results with different seed
     Given <n_slots> slots
+    And   1.0 temperature
     Then  the server is starting
     Then  the server is healthy
 

examples/server/tests/features/steps/steps.py

@@ -199,7 +199,7 @@ async def step_wait_for_the_server_to_be_started(context, expecting_status):
 
         case 'ready' | 'idle':
             await wait_for_health_status(context, context.base_url, 200, 'ok',
-                                         timeout=10,
+                                         timeout=30,
                                          params={'fail_on_no_slot': 0, 'include_slots': 0},
                                          slots_idle=context.n_slots,
                                          slots_processing=0,
@@ -883,7 +883,7 @@ async def request_completion(prompt,
                                     "cache_prompt": cache_prompt,
                                     "id_slot": id_slot,
                                     "seed": seed if seed is not None else 42,
-                                    "temperature": temperature if temperature is not None else "0.8f",
+                                    "temperature": temperature if temperature is not None else 0.8,
                                     "n_probs": 2,
                                 },
                                 headers=headers,