Add Nemotron-120B Claude Code results for all 3 benchmarks

results/nemotron-120b-rh-swe-bench-claude-code.json

@@ -0,0 +1,48 @@
+{
+    "benchmark": {
+        "name": "RH SWE-Bench",
+        "repo": "rounakbende10/rh-swe-bench",
+        "num_tasks": 357,
+        "url": "https://huggingface.co/datasets/rounakbende/rh-swe-bench"
+    },
+    "harness": {
+        "name": "Claude Code",
+        "skills": [],
+        "is_oss": false,
+        "url": "https://github.com/anthropics/claude-code"
+    },
+    "model": {
+        "name": "Nemotron-3-Super-120B-NVFP4",
+        "repo": "RedHatAI/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4",
+        "is_oss": true,
+        "num_params": 120,
+        "precision": "nvfp4",
+        "url": "https://huggingface.co/RedHatAI/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4"
+    },
+    "environment": {
+        "name": "harbor",
+        "config": {
+            "name": "rh-swe-bench"
+        },
+        "url": "https://github.com/harbor-framework/harbor"
+    },
+    "metrics": {
+        "n_tasks": 357,
+        "n_errors": 14,
+        "score": 0.224,
+        "n_input_tokens": 783670817,
+        "n_cache_tokens": 0,
+        "n_output_tokens": 5510291,
+        "n_total_tokens": 789181108,
+        "agent_time_seconds": 144352,
+        "total_time_seconds": 355122,
+        "cost_usd": 38.03,
+        "mean_input_tokens_per_task": 2195156,
+        "mean_cache_tokens_per_task": 0,
+        "mean_output_tokens_per_task": 15434,
+        "mean_tokens_per_task": 2210591,
+        "mean_cost_usd_per_task": 0.11,
+        "mean_total_time_seconds_per_task": 994,
+        "mean_agent_time_seconds_per_task": 404
+    }
+}

results/nemotron-120b-swebench-pro-ansible-claude-code.json

@@ -0,0 +1,48 @@
+{
+    "benchmark": {
+        "name": "SWE-Bench Pro -- Ansible",
+        "repo": "ScaleAI/SWE-bench_Pro",
+        "num_tasks": 96,
+        "url": "https://huggingface.co/datasets/ScaleAI/SWE-bench_Pro"
+    },
+    "harness": {
+        "name": "Claude Code",
+        "skills": [],
+        "is_oss": false,
+        "url": "https://github.com/anthropics/claude-code"
+    },
+    "model": {
+        "name": "Nemotron-3-Super-120B-NVFP4",
+        "repo": "RedHatAI/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4",
+        "is_oss": true,
+        "num_params": 120,
+        "precision": "nvfp4",
+        "url": "https://huggingface.co/RedHatAI/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4"
+    },
+    "environment": {
+        "name": "harbor",
+        "config": {
+            "name": "swe-bench-pro-ansible"
+        },
+        "url": "https://github.com/harbor-framework/harbor"
+    },
+    "metrics": {
+        "n_tasks": 96,
+        "n_errors": 0,
+        "score": 0.432,
+        "n_input_tokens": 406026150,
+        "n_cache_tokens": 0,
+        "n_output_tokens": 2839236,
+        "n_total_tokens": 408865386,
+        "agent_time_seconds": 100355,
+        "total_time_seconds": 115220,
+        "cost_usd": 26.44,
+        "mean_input_tokens_per_task": 4229439,
+        "mean_cache_tokens_per_task": 0,
+        "mean_output_tokens_per_task": 29575,
+        "mean_tokens_per_task": 4259014,
+        "mean_cost_usd_per_task": 0.28,
+        "mean_total_time_seconds_per_task": 1200,
+        "mean_agent_time_seconds_per_task": 1045
+    }
+}

results/nemotron-120b-swebench-verified-claude-code.json

@@ -0,0 +1,48 @@
+{
+    "benchmark": {
+        "name": "SWE-Bench Verified",
+        "repo": "SWE-bench/SWE-bench_Verified",
+        "num_tasks": 500,
+        "url": "https://huggingface.co/datasets/SWE-bench/SWE-bench_Verified"
+    },
+    "harness": {
+        "name": "Claude Code",
+        "skills": [],
+        "is_oss": false,
+        "url": "https://github.com/anthropics/claude-code"
+    },
+    "model": {
+        "name": "Nemotron-3-Super-120B-NVFP4",
+        "repo": "RedHatAI/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4",
+        "is_oss": true,
+        "num_params": 120,
+        "precision": "nvfp4",
+        "url": "https://huggingface.co/RedHatAI/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4"
+    },
+    "environment": {
+        "name": "harbor",
+        "config": {
+            "name": "swe-bench-verified"
+        },
+        "url": "https://github.com/harbor-framework/harbor"
+    },
+    "metrics": {
+        "n_tasks": 500,
+        "n_errors": 0,
+        "score": 0.472,
+        "n_input_tokens": 882330271,
+        "n_cache_tokens": 0,
+        "n_output_tokens": 6689285,
+        "n_total_tokens": 889019556,
+        "agent_time_seconds": 174288,
+        "total_time_seconds": 410828,
+        "cost_usd": 45.91,
+        "mean_input_tokens_per_task": 1764660,
+        "mean_cache_tokens_per_task": 0,
+        "mean_output_tokens_per_task": 13378,
+        "mean_tokens_per_task": 1778039,
+        "mean_cost_usd_per_task": 0.09,
+        "mean_total_time_seconds_per_task": 821,
+        "mean_agent_time_seconds_per_task": 348
+    }
+}