Update

results.py

@@ -27,6 +27,7 @@ instance_type_mappings = {
     "c6i.2xlarge": {"cloud": "AWS", "gpu": "-", "gpuRAM": "-"},
     "c6i.4xlarge": {"cloud": "AWS", "gpu": "-", "gpuRAM": "-"},
     "c7i.4xlarge": {"cloud": "AWS", "gpu": "-", "gpuRAM": "-"},
+    "inf2.24xlarge": {"cloud": "AWS", "gpu": "12xNeuronCore v2", "gpuRAM": "192 GB"},
     "inf2.48xlarge": {"cloud": "AWS", "gpu": "24xNeuronCore v2", "gpuRAM": "384 GB"},
     "inf2.*": {"cloud": "AWS", "gpu": "-", "gpuRAM": "-"},
 }

results_arcee_supernova.py

@@ -12,6 +12,25 @@ results_arcee_supernova = {
             "tokensPerSecond": "33",
             "notes": "",
         },
+        {
+            "instanceType": "inf2.24xlarge",
+            "configurations": [
+                {
+                    "quantization": "none",
+                    "container": "transformers-neuronx",
+                    "status": "KO",
+                    "tokensPerSecond": "-",
+                    "notes": "OOM bs=2,seqlen=4096",
+                },
+                {
+                    "quantization": "none",
+                    "container": "transformers-neuronx",
+                    "status": "KO",
+                    "tokensPerSecond": "28.4",
+                    "notes": "OOM bs=2,seqlen=2048",
+                },
+            ],
+        },
         {
             "instanceType": "inf2.48xlarge",
             "quantization": "none",