diff --git a/dev/bench/data.js b/dev/bench/data.js index 4104bb2..8bb660d 100644 --- a/dev/bench/data.js +++ b/dev/bench/data.js @@ -1,54 +1,8 @@ window.BENCHMARK_DATA = { - "lastUpdate": 1723861718877, + "lastUpdate": 1723863096569, "repoUrl": "https://github.com/neuralmagic/nm-vllm-ent", "entries": { "smaller_is_better": [ - { - "commit": { - "author": { - "name": "dhuangnm", - "username": "dhuangnm", - "email": "74931910+dhuangnm@users.noreply.github.com" - }, - "committer": { - "name": "GitHub", - "username": "web-flow", - "email": "noreply@github.com" - }, - "id": "a6dbc651a8dcd91be70465c2b437dbb242422de4", - "message": "Fix nightly failures in lm-eval tests (#37)\n\nSUMMARY:\r\n- Added max_model_len to model config and updated script to use it if\r\nprovided\r\n- Break up model list files based on gpus the model runs on\r\n- Fixed nightly failures due to lower metrics than expected\r\n\r\nTEST PLAN:\r\nAll tests\r\n\r\n---------\r\n\r\nCo-authored-by: dhuangnm ", - "timestamp": "2024-07-25T21:54:10Z", - "url": "https://github.com/neuralmagic/nm-vllm-ent/commit/a6dbc651a8dcd91be70465c2b437dbb242422de4" - }, - "date": 1722047023865, - "tool": "customSmallerIsBetter", - "benches": [ - { - "name": "{\"name\": \"mean_ttft_ms\", \"description\": \"VLLM Serving - Dense\\nmodel - facebook/opt-350m\\nmax-model-len - 2048\\nsparsity - None\\nbenchmark_serving {\\n \\\"nr-qps-pair_\\\": \\\"300,1\\\",\\n \\\"dataset\\\": \\\"sharegpt\\\"\\n}\", \"gpu_description\": \"NVIDIA L4 x 1\", \"vllm_version\": \"0.5.2.0\", \"python_version\": \"3.10.12 (main, Jun 7 2023, 13:43:11) [GCC 11.3.0]\", \"torch_version\": \"2.3.0+cu121\"}", - "value": 23.053652803329214, - "unit": "ms", - "extra": "{\n \"description\": \"VLLM Serving - Dense\\nmodel - facebook/opt-350m\\nmax-model-len - 2048\\nsparsity - None\\nbenchmark_serving {\\n \\\"nr-qps-pair_\\\": \\\"300,1\\\",\\n \\\"dataset\\\": \\\"sharegpt\\\"\\n}\",\n \"benchmarking_context\": {\n \"vllm_version\": \"0.5.2.0\",\n \"python_version\": \"3.10.12 (main, Jun 7 2023, 13:43:11) [GCC 11.3.0]\",\n \"torch_version\": \"2.3.0+cu121\",\n \"torch_cuda_version\": \"12.1\",\n \"cuda_devices\": \"[_CudaDeviceProperties(name='NVIDIA L4', major=8, minor=9, total_memory=22478MB, multi_processor_count=58)]\",\n \"cuda_device_names\": [\n \"NVIDIA L4\"\n ]\n },\n \"gpu_description\": \"NVIDIA L4 x 1\",\n \"script_name\": \"benchmark_serving.py\",\n \"script_args\": {\n \"description\": \"VLLM Serving - Dense\\nmodel - facebook/opt-350m\\nmax-model-len - 2048\\nsparsity - None\\nbenchmark_serving {\\n \\\"nr-qps-pair_\\\": \\\"300,1\\\",\\n \\\"dataset\\\": \\\"sharegpt\\\"\\n}\",\n \"backend\": \"vllm\",\n \"version\": \"N/A\",\n \"base_url\": null,\n \"host\": \"127.0.0.1\",\n \"port\": 9000,\n \"endpoint\": \"/generate\",\n \"dataset\": \"sharegpt\",\n \"num_input_tokens\": null,\n \"num_output_tokens\": null,\n \"model\": \"facebook/opt-350m\",\n \"tokenizer\": \"facebook/opt-350m\",\n \"best_of\": 1,\n \"use_beam_search\": false,\n \"log_model_io\": false,\n \"seed\": 0,\n \"trust_remote_code\": false,\n \"disable_tqdm\": false,\n \"save_directory\": \"benchmark-results\",\n \"num_prompts_\": null,\n \"request_rate_\": null,\n \"nr_qps_pair_\": [\n 300,\n \"1.0\"\n ],\n \"server_tensor_parallel_size\": 1,\n \"server_args\": \"{'model': 'facebook/opt-350m', 'tokenizer': 'facebook/opt-350m', 'max-model-len': 2048, 'host': '127.0.0.1', 'port': 9000, 'tensor-parallel-size': 1, 'disable-log-requests': ''}\"\n },\n \"date\": \"2024-07-27 02:14:40 UTC\",\n \"model\": \"facebook/opt-350m\",\n \"dataset\": \"sharegpt\"\n}" - }, - { - "name": "{\"name\": \"mean_tpot_ms\", \"description\": \"VLLM Serving - Dense\\nmodel - facebook/opt-350m\\nmax-model-len - 2048\\nsparsity - None\\nbenchmark_serving {\\n \\\"nr-qps-pair_\\\": \\\"300,1\\\",\\n \\\"dataset\\\": \\\"sharegpt\\\"\\n}\", \"gpu_description\": \"NVIDIA L4 x 1\", \"vllm_version\": \"0.5.2.0\", \"python_version\": \"3.10.12 (main, Jun 7 2023, 13:43:11) [GCC 11.3.0]\", \"torch_version\": \"2.3.0+cu121\"}", - "value": 6.059521721943343, - "unit": "ms", - "extra": "{\n \"description\": \"VLLM Serving - Dense\\nmodel - facebook/opt-350m\\nmax-model-len - 2048\\nsparsity - None\\nbenchmark_serving {\\n \\\"nr-qps-pair_\\\": \\\"300,1\\\",\\n \\\"dataset\\\": \\\"sharegpt\\\"\\n}\",\n \"benchmarking_context\": {\n \"vllm_version\": \"0.5.2.0\",\n \"python_version\": \"3.10.12 (main, Jun 7 2023, 13:43:11) [GCC 11.3.0]\",\n \"torch_version\": \"2.3.0+cu121\",\n \"torch_cuda_version\": \"12.1\",\n \"cuda_devices\": \"[_CudaDeviceProperties(name='NVIDIA L4', major=8, minor=9, total_memory=22478MB, multi_processor_count=58)]\",\n \"cuda_device_names\": [\n \"NVIDIA L4\"\n ]\n },\n \"gpu_description\": \"NVIDIA L4 x 1\",\n \"script_name\": \"benchmark_serving.py\",\n \"script_args\": {\n \"description\": \"VLLM Serving - Dense\\nmodel - facebook/opt-350m\\nmax-model-len - 2048\\nsparsity - None\\nbenchmark_serving {\\n \\\"nr-qps-pair_\\\": \\\"300,1\\\",\\n \\\"dataset\\\": \\\"sharegpt\\\"\\n}\",\n \"backend\": \"vllm\",\n \"version\": \"N/A\",\n \"base_url\": null,\n \"host\": \"127.0.0.1\",\n \"port\": 9000,\n \"endpoint\": \"/generate\",\n \"dataset\": \"sharegpt\",\n \"num_input_tokens\": null,\n \"num_output_tokens\": null,\n \"model\": \"facebook/opt-350m\",\n \"tokenizer\": \"facebook/opt-350m\",\n \"best_of\": 1,\n \"use_beam_search\": false,\n \"log_model_io\": false,\n \"seed\": 0,\n \"trust_remote_code\": false,\n \"disable_tqdm\": false,\n \"save_directory\": \"benchmark-results\",\n \"num_prompts_\": null,\n \"request_rate_\": null,\n \"nr_qps_pair_\": [\n 300,\n \"1.0\"\n ],\n \"server_tensor_parallel_size\": 1,\n \"server_args\": \"{'model': 'facebook/opt-350m', 'tokenizer': 'facebook/opt-350m', 'max-model-len': 2048, 'host': '127.0.0.1', 'port': 9000, 'tensor-parallel-size': 1, 'disable-log-requests': ''}\"\n },\n \"date\": \"2024-07-27 02:14:40 UTC\",\n \"model\": \"facebook/opt-350m\",\n \"dataset\": \"sharegpt\"\n}" - }, - { - "name": "{\"name\": \"mean_ttft_ms\", \"description\": \"VLLM Serving - Dense\\nmodel - meta-llama/Meta-Llama-3-8B-Instruct\\nmax-model-len - 4096\\nsparsity - None\\nbenchmark_serving {\\n \\\"nr-qps-pair_\\\": \\\"300,1\\\",\\n \\\"dataset\\\": \\\"sharegpt\\\"\\n}\", \"gpu_description\": \"NVIDIA L4 x 1\", \"vllm_version\": \"0.5.2.0\", \"python_version\": \"3.10.12 (main, Jun 7 2023, 13:43:11) [GCC 11.3.0]\", \"torch_version\": \"2.3.0+cu121\"}", - "value": 186.65583837666418, - "unit": "ms", - "extra": "{\n \"description\": \"VLLM Serving - Dense\\nmodel - meta-llama/Meta-Llama-3-8B-Instruct\\nmax-model-len - 4096\\nsparsity - None\\nbenchmark_serving {\\n \\\"nr-qps-pair_\\\": \\\"300,1\\\",\\n \\\"dataset\\\": \\\"sharegpt\\\"\\n}\",\n \"benchmarking_context\": {\n \"vllm_version\": \"0.5.2.0\",\n \"python_version\": \"3.10.12 (main, Jun 7 2023, 13:43:11) [GCC 11.3.0]\",\n \"torch_version\": \"2.3.0+cu121\",\n \"torch_cuda_version\": \"12.1\",\n \"cuda_devices\": \"[_CudaDeviceProperties(name='NVIDIA L4', major=8, minor=9, total_memory=22478MB, multi_processor_count=58)]\",\n \"cuda_device_names\": [\n \"NVIDIA L4\"\n ]\n },\n \"gpu_description\": \"NVIDIA L4 x 1\",\n \"script_name\": \"benchmark_serving.py\",\n \"script_args\": {\n \"description\": \"VLLM Serving - Dense\\nmodel - meta-llama/Meta-Llama-3-8B-Instruct\\nmax-model-len - 4096\\nsparsity - None\\nbenchmark_serving {\\n \\\"nr-qps-pair_\\\": \\\"300,1\\\",\\n \\\"dataset\\\": \\\"sharegpt\\\"\\n}\",\n \"backend\": \"vllm\",\n \"version\": \"N/A\",\n \"base_url\": null,\n \"host\": \"127.0.0.1\",\n \"port\": 9000,\n \"endpoint\": \"/generate\",\n \"dataset\": \"sharegpt\",\n \"num_input_tokens\": null,\n \"num_output_tokens\": null,\n \"model\": \"meta-llama/Meta-Llama-3-8B-Instruct\",\n \"tokenizer\": \"meta-llama/Meta-Llama-3-8B-Instruct\",\n \"best_of\": 1,\n \"use_beam_search\": false,\n \"log_model_io\": false,\n \"seed\": 0,\n \"trust_remote_code\": false,\n \"disable_tqdm\": false,\n \"save_directory\": \"benchmark-results\",\n \"num_prompts_\": null,\n \"request_rate_\": null,\n \"nr_qps_pair_\": [\n 300,\n \"1.0\"\n ],\n \"server_tensor_parallel_size\": 1,\n \"server_args\": \"{'model': 'meta-llama/Meta-Llama-3-8B-Instruct', 'tokenizer': 'meta-llama/Meta-Llama-3-8B-Instruct', 'max-model-len': 4096, 'host': '127.0.0.1', 'port': 9000, 'tensor-parallel-size': 1, 'disable-log-requests': ''}\"\n },\n \"date\": \"2024-07-27 02:22:36 UTC\",\n \"model\": \"meta-llama/Meta-Llama-3-8B-Instruct\",\n \"dataset\": \"sharegpt\"\n}" - }, - { - "name": "{\"name\": \"mean_tpot_ms\", \"description\": \"VLLM Serving - Dense\\nmodel - meta-llama/Meta-Llama-3-8B-Instruct\\nmax-model-len - 4096\\nsparsity - None\\nbenchmark_serving {\\n \\\"nr-qps-pair_\\\": \\\"300,1\\\",\\n \\\"dataset\\\": \\\"sharegpt\\\"\\n}\", \"gpu_description\": \"NVIDIA L4 x 1\", \"vllm_version\": \"0.5.2.0\", \"python_version\": \"3.10.12 (main, Jun 7 2023, 13:43:11) [GCC 11.3.0]\", \"torch_version\": \"2.3.0+cu121\"}", - "value": 84.26169617375099, - "unit": "ms", - "extra": "{\n \"description\": \"VLLM Serving - Dense\\nmodel - meta-llama/Meta-Llama-3-8B-Instruct\\nmax-model-len - 4096\\nsparsity - None\\nbenchmark_serving {\\n \\\"nr-qps-pair_\\\": \\\"300,1\\\",\\n \\\"dataset\\\": \\\"sharegpt\\\"\\n}\",\n \"benchmarking_context\": {\n \"vllm_version\": \"0.5.2.0\",\n \"python_version\": \"3.10.12 (main, Jun 7 2023, 13:43:11) [GCC 11.3.0]\",\n \"torch_version\": \"2.3.0+cu121\",\n \"torch_cuda_version\": \"12.1\",\n \"cuda_devices\": \"[_CudaDeviceProperties(name='NVIDIA L4', major=8, minor=9, total_memory=22478MB, multi_processor_count=58)]\",\n \"cuda_device_names\": [\n \"NVIDIA L4\"\n ]\n },\n \"gpu_description\": \"NVIDIA L4 x 1\",\n \"script_name\": \"benchmark_serving.py\",\n \"script_args\": {\n \"description\": \"VLLM Serving - Dense\\nmodel - meta-llama/Meta-Llama-3-8B-Instruct\\nmax-model-len - 4096\\nsparsity - None\\nbenchmark_serving {\\n \\\"nr-qps-pair_\\\": \\\"300,1\\\",\\n \\\"dataset\\\": \\\"sharegpt\\\"\\n}\",\n \"backend\": \"vllm\",\n \"version\": \"N/A\",\n \"base_url\": null,\n \"host\": \"127.0.0.1\",\n \"port\": 9000,\n \"endpoint\": \"/generate\",\n \"dataset\": \"sharegpt\",\n \"num_input_tokens\": null,\n \"num_output_tokens\": null,\n \"model\": \"meta-llama/Meta-Llama-3-8B-Instruct\",\n \"tokenizer\": \"meta-llama/Meta-Llama-3-8B-Instruct\",\n \"best_of\": 1,\n \"use_beam_search\": false,\n \"log_model_io\": false,\n \"seed\": 0,\n \"trust_remote_code\": false,\n \"disable_tqdm\": false,\n \"save_directory\": \"benchmark-results\",\n \"num_prompts_\": null,\n \"request_rate_\": null,\n \"nr_qps_pair_\": [\n 300,\n \"1.0\"\n ],\n \"server_tensor_parallel_size\": 1,\n \"server_args\": \"{'model': 'meta-llama/Meta-Llama-3-8B-Instruct', 'tokenizer': 'meta-llama/Meta-Llama-3-8B-Instruct', 'max-model-len': 4096, 'host': '127.0.0.1', 'port': 9000, 'tensor-parallel-size': 1, 'disable-log-requests': ''}\"\n },\n \"date\": \"2024-07-27 02:22:36 UTC\",\n \"model\": \"meta-llama/Meta-Llama-3-8B-Instruct\",\n \"dataset\": \"sharegpt\"\n}" - } - ] - }, { "commit": { "author": { @@ -2290,6 +2244,52 @@ window.BENCHMARK_DATA = { "extra": "{\n \"description\": \"VLLM Serving - Dense\\nmodel - meta-llama/Meta-Llama-3-70B-Instruct\\nmax-model-len - 4096\\nsparsity - None\\nbenchmark_serving {\\n \\\"nr-qps-pair_\\\": \\\"300,1\\\",\\n \\\"dataset\\\": \\\"sharegpt\\\"\\n}\",\n \"benchmarking_context\": {\n \"vllm_version\": \"0.5.3.0\",\n \"python_version\": \"3.10.12 (main, Jun 7 2023, 13:43:11) [GCC 11.3.0]\",\n \"torch_version\": \"2.3.1+cu121\",\n \"torch_cuda_version\": \"12.1\",\n \"cuda_devices\": \"[_CudaDeviceProperties(name='NVIDIA H100 80GB HBM3', major=9, minor=0, total_memory=81116MB, multi_processor_count=132), _CudaDeviceProperties(name='NVIDIA H100 80GB HBM3', major=9, minor=0, total_memory=81116MB, multi_processor_count=132), _CudaDeviceProperties(name='NVIDIA H100 80GB HBM3', major=9, minor=0, total_memory=81116MB, multi_processor_count=132), _CudaDeviceProperties(name='NVIDIA H100 80GB HBM3', major=9, minor=0, total_memory=81116MB, multi_processor_count=132)]\",\n \"cuda_device_names\": [\n \"NVIDIA H100 80GB HBM3\",\n \"NVIDIA H100 80GB HBM3\",\n \"NVIDIA H100 80GB HBM3\",\n \"NVIDIA H100 80GB HBM3\"\n ]\n },\n \"gpu_description\": \"NVIDIA H100 80GB HBM3 x 4\",\n \"script_name\": \"benchmark_serving.py\",\n \"script_args\": {\n \"description\": \"VLLM Serving - Dense\\nmodel - meta-llama/Meta-Llama-3-70B-Instruct\\nmax-model-len - 4096\\nsparsity - None\\nbenchmark_serving {\\n \\\"nr-qps-pair_\\\": \\\"300,1\\\",\\n \\\"dataset\\\": \\\"sharegpt\\\"\\n}\",\n \"backend\": \"vllm\",\n \"version\": \"N/A\",\n \"base_url\": null,\n \"host\": \"127.0.0.1\",\n \"port\": 9000,\n \"endpoint\": \"/generate\",\n \"dataset\": \"sharegpt\",\n \"num_input_tokens\": null,\n \"num_output_tokens\": null,\n \"model\": \"meta-llama/Meta-Llama-3-70B-Instruct\",\n \"tokenizer\": \"meta-llama/Meta-Llama-3-70B-Instruct\",\n \"best_of\": 1,\n \"use_beam_search\": false,\n \"log_model_io\": false,\n \"seed\": 0,\n \"trust_remote_code\": false,\n \"disable_tqdm\": false,\n \"save_directory\": \"benchmark-results\",\n \"num_prompts_\": null,\n \"request_rate_\": null,\n \"nr_qps_pair_\": [\n 300,\n \"1.0\"\n ],\n \"server_tensor_parallel_size\": 4,\n \"server_args\": \"{'model': 'meta-llama/Meta-Llama-3-70B-Instruct', 'tokenizer': 'meta-llama/Meta-Llama-3-70B-Instruct', 'max-model-len': 4096, 'host': '127.0.0.1', 'port': 9000, 'tensor-parallel-size': 4, 'disable-log-requests': ''}\"\n },\n \"date\": \"2024-08-17 02:27:21 UTC\",\n \"model\": \"meta-llama/Meta-Llama-3-70B-Instruct\",\n \"dataset\": \"sharegpt\"\n}" } ] + }, + { + "commit": { + "author": { + "name": "Domenic Barbuzzi", + "username": "dbarbuzzi", + "email": "domenic@neuralmagic.com" + }, + "committer": { + "name": "GitHub", + "username": "web-flow", + "email": "noreply@github.com" + }, + "id": "2efcd90ff72f4ffe8637f118ad2844d9e44bf789", + "message": "Extended server benchmark model list (#40)\n\nThis PR updates the benchmark configs to include some a large model that\r\nrun on the 4x H100 runner.\r\n\r\nTo accomplish this, the various `benchmark_` inputs to `build-test` are\r\nconsolidated into the JSON-based `benchmark_configs` input (which\r\naccepts a list of objects containing keys: `label`, `config_file`,\r\n`timeout`).\r\n\r\nThis allows the existing benchmarking of ‘small’ models to continue\r\nrunning on the runners they already are, while allowing a new config\r\nwith ‘large’ models to run specifically on the 4x H100 machine.\r\n\r\nAdditionally, the config now supports an optionally-present property to\r\ndefine the server startup timeout, which is set on the large model\r\nconfig (60 min). This will override the default (15 min) if present. In\r\nthe course of adding this support, I fixed a handful of incorrect type\r\nhinting about the config object (as it tripped me up when I tried to use\r\nit as the advertised type, which didn’t work).", + "timestamp": "2024-08-16T15:16:47Z", + "url": "https://github.com/neuralmagic/nm-vllm-ent/commit/2efcd90ff72f4ffe8637f118ad2844d9e44bf789" + }, + "date": 1723863096249, + "tool": "customSmallerIsBetter", + "benches": [ + { + "name": "{\"name\": \"mean_ttft_ms\", \"description\": \"VLLM Serving - Dense\\nmodel - meta-llama/Meta-Llama-3-8B-Instruct\\nmax-model-len - 4096\\nsparsity - None\\nbenchmark_serving {\\n \\\"nr-qps-pair_\\\": \\\"300,1\\\",\\n \\\"dataset\\\": \\\"sharegpt\\\"\\n}\", \"gpu_description\": \"NVIDIA H100 80GB HBM3 x 1\", \"vllm_version\": \"0.5.3.0\", \"python_version\": \"3.10.12 (main, Jun 7 2023, 13:43:11) [GCC 11.3.0]\", \"torch_version\": \"2.3.1+cu121\"}", + "value": 37.40427045151591, + "unit": "ms", + "extra": "{\n \"description\": \"VLLM Serving - Dense\\nmodel - meta-llama/Meta-Llama-3-8B-Instruct\\nmax-model-len - 4096\\nsparsity - None\\nbenchmark_serving {\\n \\\"nr-qps-pair_\\\": \\\"300,1\\\",\\n \\\"dataset\\\": \\\"sharegpt\\\"\\n}\",\n \"benchmarking_context\": {\n \"vllm_version\": \"0.5.3.0\",\n \"python_version\": \"3.10.12 (main, Jun 7 2023, 13:43:11) [GCC 11.3.0]\",\n \"torch_version\": \"2.3.1+cu121\",\n \"torch_cuda_version\": \"12.1\",\n \"cuda_devices\": \"[_CudaDeviceProperties(name='NVIDIA H100 80GB HBM3', major=9, minor=0, total_memory=81116MB, multi_processor_count=132)]\",\n \"cuda_device_names\": [\n \"NVIDIA H100 80GB HBM3\"\n ]\n },\n \"gpu_description\": \"NVIDIA H100 80GB HBM3 x 1\",\n \"script_name\": \"benchmark_serving.py\",\n \"script_args\": {\n \"description\": \"VLLM Serving - Dense\\nmodel - meta-llama/Meta-Llama-3-8B-Instruct\\nmax-model-len - 4096\\nsparsity - None\\nbenchmark_serving {\\n \\\"nr-qps-pair_\\\": \\\"300,1\\\",\\n \\\"dataset\\\": \\\"sharegpt\\\"\\n}\",\n \"backend\": \"vllm\",\n \"version\": \"N/A\",\n \"base_url\": null,\n \"host\": \"127.0.0.1\",\n \"port\": 9000,\n \"endpoint\": \"/generate\",\n \"dataset\": \"sharegpt\",\n \"num_input_tokens\": null,\n \"num_output_tokens\": null,\n \"model\": \"meta-llama/Meta-Llama-3-8B-Instruct\",\n \"tokenizer\": \"meta-llama/Meta-Llama-3-8B-Instruct\",\n \"best_of\": 1,\n \"use_beam_search\": false,\n \"log_model_io\": false,\n \"seed\": 0,\n \"trust_remote_code\": false,\n \"disable_tqdm\": false,\n \"save_directory\": \"benchmark-results\",\n \"num_prompts_\": null,\n \"request_rate_\": null,\n \"nr_qps_pair_\": [\n 300,\n \"1.0\"\n ],\n \"server_tensor_parallel_size\": 1,\n \"server_args\": \"{'model': 'meta-llama/Meta-Llama-3-8B-Instruct', 'tokenizer': 'meta-llama/Meta-Llama-3-8B-Instruct', 'max-model-len': 4096, 'host': '127.0.0.1', 'port': 9000, 'tensor-parallel-size': 1, 'disable-log-requests': ''}\"\n },\n \"date\": \"2024-08-17 02:50:13 UTC\",\n \"model\": \"meta-llama/Meta-Llama-3-8B-Instruct\",\n \"dataset\": \"sharegpt\"\n}" + }, + { + "name": "{\"name\": \"mean_tpot_ms\", \"description\": \"VLLM Serving - Dense\\nmodel - meta-llama/Meta-Llama-3-8B-Instruct\\nmax-model-len - 4096\\nsparsity - None\\nbenchmark_serving {\\n \\\"nr-qps-pair_\\\": \\\"300,1\\\",\\n \\\"dataset\\\": \\\"sharegpt\\\"\\n}\", \"gpu_description\": \"NVIDIA H100 80GB HBM3 x 1\", \"vllm_version\": \"0.5.3.0\", \"python_version\": \"3.10.12 (main, Jun 7 2023, 13:43:11) [GCC 11.3.0]\", \"torch_version\": \"2.3.1+cu121\"}", + "value": 12.310676973138039, + "unit": "ms", + "extra": "{\n \"description\": \"VLLM Serving - Dense\\nmodel - meta-llama/Meta-Llama-3-8B-Instruct\\nmax-model-len - 4096\\nsparsity - None\\nbenchmark_serving {\\n \\\"nr-qps-pair_\\\": \\\"300,1\\\",\\n \\\"dataset\\\": \\\"sharegpt\\\"\\n}\",\n \"benchmarking_context\": {\n \"vllm_version\": \"0.5.3.0\",\n \"python_version\": \"3.10.12 (main, Jun 7 2023, 13:43:11) [GCC 11.3.0]\",\n \"torch_version\": \"2.3.1+cu121\",\n \"torch_cuda_version\": \"12.1\",\n \"cuda_devices\": \"[_CudaDeviceProperties(name='NVIDIA H100 80GB HBM3', major=9, minor=0, total_memory=81116MB, multi_processor_count=132)]\",\n \"cuda_device_names\": [\n \"NVIDIA H100 80GB HBM3\"\n ]\n },\n \"gpu_description\": \"NVIDIA H100 80GB HBM3 x 1\",\n \"script_name\": \"benchmark_serving.py\",\n \"script_args\": {\n \"description\": \"VLLM Serving - Dense\\nmodel - meta-llama/Meta-Llama-3-8B-Instruct\\nmax-model-len - 4096\\nsparsity - None\\nbenchmark_serving {\\n \\\"nr-qps-pair_\\\": \\\"300,1\\\",\\n \\\"dataset\\\": \\\"sharegpt\\\"\\n}\",\n \"backend\": \"vllm\",\n \"version\": \"N/A\",\n \"base_url\": null,\n \"host\": \"127.0.0.1\",\n \"port\": 9000,\n \"endpoint\": \"/generate\",\n \"dataset\": \"sharegpt\",\n \"num_input_tokens\": null,\n \"num_output_tokens\": null,\n \"model\": \"meta-llama/Meta-Llama-3-8B-Instruct\",\n \"tokenizer\": \"meta-llama/Meta-Llama-3-8B-Instruct\",\n \"best_of\": 1,\n \"use_beam_search\": false,\n \"log_model_io\": false,\n \"seed\": 0,\n \"trust_remote_code\": false,\n \"disable_tqdm\": false,\n \"save_directory\": \"benchmark-results\",\n \"num_prompts_\": null,\n \"request_rate_\": null,\n \"nr_qps_pair_\": [\n 300,\n \"1.0\"\n ],\n \"server_tensor_parallel_size\": 1,\n \"server_args\": \"{'model': 'meta-llama/Meta-Llama-3-8B-Instruct', 'tokenizer': 'meta-llama/Meta-Llama-3-8B-Instruct', 'max-model-len': 4096, 'host': '127.0.0.1', 'port': 9000, 'tensor-parallel-size': 1, 'disable-log-requests': ''}\"\n },\n \"date\": \"2024-08-17 02:50:13 UTC\",\n \"model\": \"meta-llama/Meta-Llama-3-8B-Instruct\",\n \"dataset\": \"sharegpt\"\n}" + }, + { + "name": "{\"name\": \"mean_ttft_ms\", \"description\": \"VLLM Serving - Dense\\nmodel - facebook/opt-350m\\nmax-model-len - 2048\\nsparsity - None\\nbenchmark_serving {\\n \\\"nr-qps-pair_\\\": \\\"300,1\\\",\\n \\\"dataset\\\": \\\"sharegpt\\\"\\n}\", \"gpu_description\": \"NVIDIA H100 80GB HBM3 x 1\", \"vllm_version\": \"0.5.3.0\", \"python_version\": \"3.10.12 (main, Jun 7 2023, 13:43:11) [GCC 11.3.0]\", \"torch_version\": \"2.3.1+cu121\"}", + "value": 42.3081368021667, + "unit": "ms", + "extra": "{\n \"description\": \"VLLM Serving - Dense\\nmodel - facebook/opt-350m\\nmax-model-len - 2048\\nsparsity - None\\nbenchmark_serving {\\n \\\"nr-qps-pair_\\\": \\\"300,1\\\",\\n \\\"dataset\\\": \\\"sharegpt\\\"\\n}\",\n \"benchmarking_context\": {\n \"vllm_version\": \"0.5.3.0\",\n \"python_version\": \"3.10.12 (main, Jun 7 2023, 13:43:11) [GCC 11.3.0]\",\n \"torch_version\": \"2.3.1+cu121\",\n \"torch_cuda_version\": \"12.1\",\n \"cuda_devices\": \"[_CudaDeviceProperties(name='NVIDIA H100 80GB HBM3', major=9, minor=0, total_memory=81116MB, multi_processor_count=132)]\",\n \"cuda_device_names\": [\n \"NVIDIA H100 80GB HBM3\"\n ]\n },\n \"gpu_description\": \"NVIDIA H100 80GB HBM3 x 1\",\n \"script_name\": \"benchmark_serving.py\",\n \"script_args\": {\n \"description\": \"VLLM Serving - Dense\\nmodel - facebook/opt-350m\\nmax-model-len - 2048\\nsparsity - None\\nbenchmark_serving {\\n \\\"nr-qps-pair_\\\": \\\"300,1\\\",\\n \\\"dataset\\\": \\\"sharegpt\\\"\\n}\",\n \"backend\": \"vllm\",\n \"version\": \"N/A\",\n \"base_url\": null,\n \"host\": \"127.0.0.1\",\n \"port\": 9000,\n \"endpoint\": \"/generate\",\n \"dataset\": \"sharegpt\",\n \"num_input_tokens\": null,\n \"num_output_tokens\": null,\n \"model\": \"facebook/opt-350m\",\n \"tokenizer\": \"facebook/opt-350m\",\n \"best_of\": 1,\n \"use_beam_search\": false,\n \"log_model_io\": false,\n \"seed\": 0,\n \"trust_remote_code\": false,\n \"disable_tqdm\": false,\n \"save_directory\": \"benchmark-results\",\n \"num_prompts_\": null,\n \"request_rate_\": null,\n \"nr_qps_pair_\": [\n 300,\n \"1.0\"\n ],\n \"server_tensor_parallel_size\": 1,\n \"server_args\": \"{'model': 'facebook/opt-350m', 'tokenizer': 'facebook/opt-350m', 'max-model-len': 2048, 'host': '127.0.0.1', 'port': 9000, 'tensor-parallel-size': 1, 'disable-log-requests': ''}\"\n },\n \"date\": \"2024-08-17 02:41:23 UTC\",\n \"model\": \"facebook/opt-350m\",\n \"dataset\": \"sharegpt\"\n}" + }, + { + "name": "{\"name\": \"mean_tpot_ms\", \"description\": \"VLLM Serving - Dense\\nmodel - facebook/opt-350m\\nmax-model-len - 2048\\nsparsity - None\\nbenchmark_serving {\\n \\\"nr-qps-pair_\\\": \\\"300,1\\\",\\n \\\"dataset\\\": \\\"sharegpt\\\"\\n}\", \"gpu_description\": \"NVIDIA H100 80GB HBM3 x 1\", \"vllm_version\": \"0.5.3.0\", \"python_version\": \"3.10.12 (main, Jun 7 2023, 13:43:11) [GCC 11.3.0]\", \"torch_version\": \"2.3.1+cu121\"}", + "value": 7.405811458789267, + "unit": "ms", + "extra": "{\n \"description\": \"VLLM Serving - Dense\\nmodel - facebook/opt-350m\\nmax-model-len - 2048\\nsparsity - None\\nbenchmark_serving {\\n \\\"nr-qps-pair_\\\": \\\"300,1\\\",\\n \\\"dataset\\\": \\\"sharegpt\\\"\\n}\",\n \"benchmarking_context\": {\n \"vllm_version\": \"0.5.3.0\",\n \"python_version\": \"3.10.12 (main, Jun 7 2023, 13:43:11) [GCC 11.3.0]\",\n \"torch_version\": \"2.3.1+cu121\",\n \"torch_cuda_version\": \"12.1\",\n \"cuda_devices\": \"[_CudaDeviceProperties(name='NVIDIA H100 80GB HBM3', major=9, minor=0, total_memory=81116MB, multi_processor_count=132)]\",\n \"cuda_device_names\": [\n \"NVIDIA H100 80GB HBM3\"\n ]\n },\n \"gpu_description\": \"NVIDIA H100 80GB HBM3 x 1\",\n \"script_name\": \"benchmark_serving.py\",\n \"script_args\": {\n \"description\": \"VLLM Serving - Dense\\nmodel - facebook/opt-350m\\nmax-model-len - 2048\\nsparsity - None\\nbenchmark_serving {\\n \\\"nr-qps-pair_\\\": \\\"300,1\\\",\\n \\\"dataset\\\": \\\"sharegpt\\\"\\n}\",\n \"backend\": \"vllm\",\n \"version\": \"N/A\",\n \"base_url\": null,\n \"host\": \"127.0.0.1\",\n \"port\": 9000,\n \"endpoint\": \"/generate\",\n \"dataset\": \"sharegpt\",\n \"num_input_tokens\": null,\n \"num_output_tokens\": null,\n \"model\": \"facebook/opt-350m\",\n \"tokenizer\": \"facebook/opt-350m\",\n \"best_of\": 1,\n \"use_beam_search\": false,\n \"log_model_io\": false,\n \"seed\": 0,\n \"trust_remote_code\": false,\n \"disable_tqdm\": false,\n \"save_directory\": \"benchmark-results\",\n \"num_prompts_\": null,\n \"request_rate_\": null,\n \"nr_qps_pair_\": [\n 300,\n \"1.0\"\n ],\n \"server_tensor_parallel_size\": 1,\n \"server_args\": \"{'model': 'facebook/opt-350m', 'tokenizer': 'facebook/opt-350m', 'max-model-len': 2048, 'host': '127.0.0.1', 'port': 9000, 'tensor-parallel-size': 1, 'disable-log-requests': ''}\"\n },\n \"date\": \"2024-08-17 02:41:23 UTC\",\n \"model\": \"facebook/opt-350m\",\n \"dataset\": \"sharegpt\"\n}" + } + ] } ] }