Add FlashHead version to measurement setup

config.json

@@ -221,9 +221,9 @@
         "fps"
       ],
       "experiment_setup": {
-        "agx_thor": "Measurement setup: NVIDIA AI IoT vLLM 0.16.0 arm64, 256 tokens generated, 10 warm-up runs, averaged over 25 runs.",
-        "agx_orin": "Measurement setup: NVIDIA AI IoT vLLM 0.16.0 tegra, 256 tokens generated, 10 warm-up runs, averaged over 25 runs.",
-        "orin_nano": "Measurement setup: NVIDIA AI IoT vLLM 0.16.0 tegra, 256 tokens generated, 10 warm-up runs, averaged over 25 runs."
+        "agx_thor": "Measurement setup: NVIDIA AI IoT vLLM 0.16.0 arm64, FlashHead 0.1.9, 256 tokens generated, 10 warm-up runs, averaged over 25 runs.",
+        "agx_orin": "Measurement setup: NVIDIA AI IoT vLLM 0.16.0 tegra, FlashHead 0.1.9, 256 tokens generated, 10 warm-up runs, averaged over 25 runs.",
+        "orin_nano": "Measurement setup: NVIDIA AI IoT vLLM 0.16.0 tegra, FlashHead 0.1.9, 256 tokens generated, 10 warm-up runs, averaged over 25 runs."
       },
       "default_device": "agx_orin"
     },
@@ -231,10 +231,10 @@
       "data_file": "data/Llama-3.2.csv",
       "accuracy_file": "data/acc-Llama-3.2.csv",
       "experiment_setup": {
-        "agx_thor": "Measurement setup: NVIDIA AI IoT vLLM 0.19.0 arm64, 32 input tokens, 256 output tokens generated, 10 warm-up runs, averaged over 25 runs.",
-        "agx_orin": "Measurement setup: NVIDIA AI IoT vLLM 0.19.0 tegra, 32 input tokens, 256 output tokens generated, 10 warm-up runs, averaged over 25 runs.",
-        "orin_nano_super": "Measurement setup: NVIDIA AI IoT vLLM 0.19.0 tegra, 32 input tokens, 256 output tokens generated, 10 warm-up runs, averaged over 25 runs.",
-        "rtx_3500_ada": "Measurement setup: vLLM 0.10.2, batch_size=1, 32 input tokens, 128 output tokens generated, 10 warm-up runs, averaged over 100 runs."
+        "agx_thor": "Measurement setup: NVIDIA AI IoT vLLM 0.19.0 arm64, FlashHead 0.1.7, 32 input tokens, 256 output tokens generated, 10 warm-up runs, averaged over 25 runs.",
+        "agx_orin": "Measurement setup: NVIDIA AI IoT vLLM 0.19.0 tegra, FlashHead 0.1.7, 32 input tokens, 256 output tokens generated, 10 warm-up runs, averaged over 25 runs.",
+        "orin_nano_super": "Measurement setup: NVIDIA AI IoT vLLM 0.19.0 tegra, FlashHead 0.1.7, 32 input tokens, 256 output tokens generated, 10 warm-up runs, averaged over 25 runs.",
+        "rtx_3500_ada": "Measurement setup: vLLM 0.10.2, FlashHead 0.1.7, batch_size=1, 32 input tokens, 128 output tokens generated, 10 warm-up runs, averaged over 100 runs."
       },
       "default_device": "agx_orin"
     },
@@ -242,10 +242,10 @@
       "data_file": "data/Gemma-3.csv",
       "accuracy_file": "data/acc-Gemma-3.csv",
       "experiment_setup": {
-        "agx_thor": "Measurement setup: NVIDIA AI IoT vLLM 0.19.0 arm64, 32 input tokens, 256 output tokens generated, 10 warm-up runs, averaged over 25 runs.",
+        "agx_thor": "Measurement setup: NVIDIA AI IoT vLLM 0.19.0 arm64, FlashHead 0.1.7, 32 input tokens, 256 output tokens generated, 10 warm-up runs, averaged over 25 runs.",
         "agx_orin": "Measurement setup: NVIDIA AI IoT vLLM 0.19.0 tegra, 32 input tokens, 256 output tokens generated, 10 warm-up runs, averaged over 25 runs.",
         "orin_nano_super": "Measurement setup: NVIDIA AI IoT vLLM 0.19.0 tegra, 32 input tokens, 256 output tokens generated, 10 warm-up runs, averaged over 25 runs.",
-        "rtx_3500_ada": "Measurement setup: vLLM 0.10.2, batch_size=1, 32 input tokens, 128 output tokens generated, 10 warm-up runs, averaged over 100 runs."
+        "rtx_3500_ada": "Measurement setup: vLLM 0.10.2, FlashHead 0.1.7, batch_size=1, 32 input tokens, 128 output tokens generated, 10 warm-up runs, averaged over 100 runs."
       },
       "default_device": "agx_orin"
     },
@@ -253,13 +253,13 @@
       "data_file": "data/Qwen3.csv",
       "accuracy_file": "data/acc-Qwen3.csv",
       "experiment_setup": {
-        "agx_thor": "Measurement setup: NVIDIA AI IoT vLLM 0.19.0 arm64, 32 input tokens, 256 output tokens generated, 10 warm-up runs, averaged over 25 runs.",
-        "agx_orin": "Measurement setup: NVIDIA AI IoT vLLM 0.19.0 tegra, 32 input tokens, 256 output tokens generated, 10 warm-up runs, averaged over 25 runs.",
+        "agx_thor": "Measurement setup: NVIDIA AI IoT vLLM 0.19.0 arm64, FlashHead 0.1.7, 32 input tokens, 256 output tokens generated, 10 warm-up runs, averaged over 25 runs.",
+        "agx_orin": "Measurement setup: NVIDIA AI IoT vLLM 0.19.0 tegra, FlashHead 0.1.7, 32 input tokens, 256 output tokens generated, 10 warm-up runs, averaged over 25 runs.",
         "orin_nano_super": "Measurement setup: NVIDIA AI IoT vLLM 0.19.0 tegra, 32 input tokens, 256 output tokens generated, 10 warm-up runs, averaged over 25 runs.",
-        "rtx_3500_ada": "Measurement setup: vLLM 0.10.2, batch_size=1, 32 input tokens, 128 output tokens generated, 10 warm-up runs, averaged over 100 runs."
+        "rtx_3500_ada": "Measurement setup: vLLM 0.10.2, FlashHead 0.1.7, batch_size=1, 32 input tokens, 128 output tokens generated, 10 warm-up runs, averaged over 100 runs."
       },
       "default_device": "agx_orin"
     }
   },
   "accuracy_title": "Accuracy"
-}
+}