Skip to content

Optimize for 2% stability on Llama-3.1-70B (8xGPU)! #47

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.

Sign up for GitHub

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

Open
BOSS10130206 wants to merge 1 commit into
base: main
Choose a base branch
from
Open

Optimize for 2% stability on Llama-3.1-70B (8xGPU)! #47

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
8 changes: 4 additions & 4 deletions alternative_recipes/helm-charts/inference/llm-benchmarks/sg-llm-server-values.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -23,23 +23,23 @@ environment:
# Server Configuration
SERVER_TYPE: "SGLang" # Can be "vLLM", "NIM", or "SGLang"
SERVER_IMAGE: "lmsysorg/sglang:latest" # Path to server image
SERVER_ARGS: ""
SERVER_ARGS: """--mem-fraction-for-model-kv-cache 0.9 --tp 8 --trust-remote-code"
SERVER_PORT: "8000" # The port the server will listen on
SERVER_HEALTH_CHECK_ENDPOINT: "health" # The endpoint to check if the server is ready
SERVER_ENDPOINT: "v1/chat/completions" # The endpoint for chat completions

# Model Configuration
MODEL_NAME: "meta-llama/Llama-3.2-1B-Instruct"
MODEL_NAME: Llama-3.1-70B-Instruct
MODEL_NAME_CLEANED: "llama-3.2-1b-instruct" # used in the benchmark result file name
MODEL_TOKENIZER: "meta-llama/Llama-3.2-1B-Instruct" # matches the path of the tokenizer on the HF registry
HF_TOKEN: "" # Your Hugging Face API token

# Benchmark Parameters
MIN_REQUESTS: "20"
USE_CASES: "chat:128/128 chat:4096/512"
CONCURRENCY_RANGE: "1 25 50 100"
CONCURRENCY_RANGE: "1 50 100 200 400"
REQUEST_MULTIPLIER: "5"
NUM_GPUS: "1" # Configure the number of GPUs to use (e.g., "1", "2", "4", "8")
NUM_GPUS: "8" # Configure the number of GPUs to use (e.g., "1", "2", "4", "8")

# Results Configuration
RESULTS_PATH: "/sg-llm-results" # Directory where benchmark results will be stored
Expand Down