{ "settings": { "allow_confirm_datapoint": true, "hide_show_config": true }, "prefill": { "chart": { "labels": [], "datasets": [ { "label": "Prefill Performance", "data": [], "backgroundColor": "#1f77b4", "borderColor": "#1f77b4" } ], "target_line": { "value": 0.0, "label": "Target TTFT: ? ms" }, "axes": { "x": { "title": "Time to First Token (ms)", "min": 0 }, "y": { "title": "Prefill Throughput per GPU (tokens/s/GPU)", "min": 0 } } }, "table": { "columns": [ "GPUs", "TTFT (ms)", "Throughput (tokens/s/GPU)", "Action" ], "data": [] } }, "decode": { "chart": { "datasets": [], "target_line": { "value": 0.0, "label": "Target ITL: ? ms" }, "axes": { "x": { "title": "Inter Token Latency (ms)", "min": 0 }, "y": { "title": "Decode Throughput per GPU (tokens/s/GPU)", "min": 0 } } }, "table": { "columns": [ "GPUs", "ITL (ms)", "Throughput (tokens/s/GPU)", "Action" ], "data": [] } }, "cost": { "chart": { "datasets": [], "axes": { "x": { "title": "Tokens per User", "min": 0 }, "y": { "title": "Cost ($)", "min": 0 } }, "title": "Cost Per 1000 ? requests" }, "table": { "columns": [ "TTFT (ms)", "Prefill Thpt (tokens/s/GPU)", "ITL (ms)", "Decode Thpt (tokens/s/GPU)", "Tokens/User", "Cost ($)", "Action" ], "data": [] } } }