Skip to content
This repository has been archived by the owner on Oct 11, 2024. It is now read-only.

Commit

Permalink
fail smaller is better job
Browse files Browse the repository at this point in the history
  • Loading branch information
Varun Sundar Rabindranath committed Mar 27, 2024
1 parent a41c281 commit 94a26cd
Showing 1 changed file with 6 additions and 0 deletions.
6 changes: 6 additions & 0 deletions .github/actions/nm-github-action-benchmark/action.yml
View file
Original file line number Diff line number Diff line change
Expand Up @@ -67,3 +67,9 @@ runs:
comment-on-alert: ${{ inputs.reporting_enabled == 'true' }}
# TODO (varun): Is this a reasonable number ?
max-items-in-chart: 50

- name: fail smaller is better job
if: ${{ inputs.gh_action_benchmark_name == 'smaller_is_better' }}
run: |
exit 1
shell: bash

3 comments on commit 94a26cd

@github-actions
Copy link

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

bigger_is_better

Benchmark suite Current: 94a26cd Previous: 8894487 Ratio
{"name": "request_throughput", "description": "Benchmark vllm serving\nmodel - mistralai/Mistral-7B-Instruct-v0.2\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"5,inf\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 4", "vllm_version": "0.1.0", "python_version": "3.10.12 (main, Mar 7 2024, 18:39:53) [GCC 9.4.0]", "torch_version": "2.1.2+cu121"} 1.0852489035865591 prompts/s
{"name": "input_throughput", "description": "Benchmark vllm serving\nmodel - mistralai/Mistral-7B-Instruct-v0.2\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"5,inf\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 4", "vllm_version": "0.1.0", "python_version": "3.10.12 (main, Mar 7 2024, 18:39:53) [GCC 9.4.0]", "torch_version": "2.1.2+cu121"} 202.0733458478173 tokens/s
{"name": "output_throughput", "description": "Benchmark vllm serving\nmodel - mistralai/Mistral-7B-Instruct-v0.2\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"5,inf\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 4", "vllm_version": "0.1.0", "python_version": "3.10.12 (main, Mar 7 2024, 18:39:53) [GCC 9.4.0]", "torch_version": "2.1.2+cu121"} 191.87200615410367 tokens/s
{"name": "request_throughput", "description": "Benchmark vllm engine throughput - with dataset\nmodel - mistralai/Mistral-7B-Instruct-v0.2\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"output-len\": 128,\n \"num-prompts\": 100,\n \"dataset\": \"sharegpt\",\n \"max-model-len\": 4096\n}", "gpu_description": "NVIDIA A10G x 4", "vllm_version": "0.1.0", "python_version": "3.10.12 (main, Mar 7 2024, 18:39:53) [GCC 9.4.0]", "torch_version": "2.1.2+cu121"} 7.8124766626673825 prompts/s
{"name": "token_throughput", "description": "Benchmark vllm engine throughput - with dataset\nmodel - mistralai/Mistral-7B-Instruct-v0.2\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"output-len\": 128,\n \"num-prompts\": 100,\n \"dataset\": \"sharegpt\",\n \"max-model-len\": 4096\n}", "gpu_description": "NVIDIA A10G x 4", "vllm_version": "0.1.0", "python_version": "3.10.12 (main, Mar 7 2024, 18:39:53) [GCC 9.4.0]", "torch_version": "2.1.2+cu121"} 3542.4112931532713 tokens/s

This comment was automatically generated by workflow using github-action-benchmark.

@github-actions
Copy link

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

smaller_is_better

Benchmark suite Current: 94a26cd Previous: 8894487 Ratio
{"name": "median_request_latency", "description": "Benchmark vllm serving\nmodel - mistralai/Mistral-7B-Instruct-v0.2\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"5,inf\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 4", "vllm_version": "0.1.0", "python_version": "3.10.12 (main, Mar 7 2024, 18:39:53) [GCC 9.4.0]", "torch_version": "2.1.2+cu121"} 3240.434912000069 ms
{"name": "mean_ttft_ms", "description": "Benchmark vllm serving\nmodel - mistralai/Mistral-7B-Instruct-v0.2\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"5,inf\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 4", "vllm_version": "0.1.0", "python_version": "3.10.12 (main, Mar 7 2024, 18:39:53) [GCC 9.4.0]", "torch_version": "2.1.2+cu121"} 237.94871639997837 ms
{"name": "median_ttft_ms", "description": "Benchmark vllm serving\nmodel - mistralai/Mistral-7B-Instruct-v0.2\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"5,inf\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 4", "vllm_version": "0.1.0", "python_version": "3.10.12 (main, Mar 7 2024, 18:39:53) [GCC 9.4.0]", "torch_version": "2.1.2+cu121"} 237.83499699993627 ms
{"name": "mean_tpot_ms", "description": "Benchmark vllm serving\nmodel - mistralai/Mistral-7B-Instruct-v0.2\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"5,inf\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 4", "vllm_version": "0.1.0", "python_version": "3.10.12 (main, Mar 7 2024, 18:39:53) [GCC 9.4.0]", "torch_version": "2.1.2+cu121"} 13.714923565798504 ms
{"name": "median_tpot_ms", "description": "Benchmark vllm serving\nmodel - mistralai/Mistral-7B-Instruct-v0.2\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"5,inf\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 4", "vllm_version": "0.1.0", "python_version": "3.10.12 (main, Mar 7 2024, 18:39:53) [GCC 9.4.0]", "torch_version": "2.1.2+cu121"} 14.25400808860743 ms

This comment was automatically generated by workflow using github-action-benchmark.

@github-actions
Copy link

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

bigger_is_better

Benchmark suite Current: 94a26cd Previous: 8894487 Ratio
{"name": "request_throughput", "description": "VLLM Engine throughput - synthetic\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 256,\n \"output-len\": 128,\n \"num-prompts\": 1000\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.1.0", "python_version": "3.10.12 (main, Mar 7 2024, 18:39:53) [GCC 9.4.0]", "torch_version": "2.1.2+cu121"} 3.9845809581855027 prompts/s 3.98509085065037 prompts/s 1.00
{"name": "token_throughput", "description": "VLLM Engine throughput - synthetic\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 256,\n \"output-len\": 128,\n \"num-prompts\": 1000\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.1.0", "python_version": "3.10.12 (main, Mar 7 2024, 18:39:53) [GCC 9.4.0]", "torch_version": "2.1.2+cu121"} 1530.0790879432332 tokens/s 1530.2748866497423 tokens/s 1.00

This comment was automatically generated by workflow using github-action-benchmark.

Please sign in to comment.