{"case": "no-schema", "latency_ms": 3943.9, "completion_tokens": 128, "valid": null, "content": "{\n  \"gpu_id\": \"NVIDIA A100 80GB\",\n  \"model\": \"NVIDIA A100\",\n  \"memory\": \"80 GB GDDR6\",\n  \"compute_capability\": \"8.0\",\n  \"power_consumption\": \"400 W\",\n  \"interfaces\": [\n    \"PCIe 4.0 x16\"\n  ],\n  \"supported_frameworks\": [\n    \"TensorFlow\",\n    \"PyTorch\",\n    \"ONNX\",\n    \"TensorRT\"\n  ],\n  \"use_case\": \"High-performance inference and training for large-scale AI"}
{"case": "new-schema", "latency_ms": 1636.6, "completion_tokens": 41, "valid": true, "content": "{\n  \"name\": \"NVIDIA A100\",\n  \"memory_gb\": 40,\n  \"vendor\": \"nvidia\",\n  \"variant\": \"v1\"\n}"}
{"case": "same-schema-again", "latency_ms": 1309.2, "completion_tokens": 41, "valid": true, "content": "{\n  \"name\": \"NVIDIA A100\",\n  \"memory_gb\": 40,\n  \"vendor\": \"nvidia\",\n  \"variant\": \"v1\"\n}"}
{"case": "same-schema-third", "latency_ms": 1310.9, "completion_tokens": 41, "valid": true, "content": "{\n  \"name\": \"NVIDIA A100\",\n  \"memory_gb\": 40,\n  \"vendor\": \"nvidia\",\n  \"variant\": \"v1\"\n}"}
{"case": "another-new-schema", "latency_ms": 1291.1, "completion_tokens": 41, "valid": true, "content": "{\n  \"name\": \"NVIDIA A100\",\n  \"memory_gb\": 40,\n  \"vendor\": \"nvidia\",\n  \"variant\": \"v2\"\n}"}