Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

fix the error of TTFT and TPOT while the bench target is chatqna_qlist_pubmed #238

Merged
merged 2 commits into from
Feb 20, 2025

Conversation

leslieluyu
Copy link
Contributor

Description

missing streaming config of chatqna_qlist_pubmed
Lack of the config will cause the error of "next token latency"

example

End to End latency(ms),    P50: 25875.62,   P90: 34988.45,   P99: 56357.57,   Avg: 26382.29
Time to First Token-TTFT(ms),   P50: 25875.21,   P90: 34987.96,   P99: 56357.17,   Avg: 26381.87
Time Per Output Token-TPOT(ms),   P50: 0.00,   P90: 0.00,   P99: 0.01,   Avg: 0.00

Issues

List the issue or RFC link this PR is working on. If there is no such link, please mark it as n/a.

Type of change

List the type of change like below. Please delete options that are not relevant.

  • Bug fix (non-breaking change which fixes an issue)

Dependencies

List the newly introduced 3rd party dependency if exists.

Tests

Describe the tests that you ran to verify your changes.

  1. deploy the chatqna v1.2
  2. git pull the evel code
    3.use locust command line to verify bench-target "chatqna_qlist_pubmed"
locust --locustfile ~/GenAIEval/evals/benchmark/stresscli/locust/aistress.py --host http://IP:PORT --run-time 60m  --max-output 128  --seed 1024 --processes 8 --users 128 --spawn-rate 100 --max-request 640 --bench-target chatqna_qlist_pubmed --llm-model meta-llama/Meta-Llama-3-8B-Instruct --stop-timeout 120 --csv /opt/opea/test_reports/test_locust/ --headless --only-summary --loglevel WARNING --json

@lvliang-intel lvliang-intel merged commit da04a9f into opea-project:main Feb 20, 2025
7 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants