28 lines
1.2 KiB
YAML
28 lines
1.2 KiB
YAML
# Basic Benchmark Configuration
|
|
apiEndpoint: "http://139.9.229.200:8080/v1/chat/completions" # Replace with your actual API endpoint
|
|
apiKey: "devserver-9aB3-7fG5q2Yd-8C4T-1R6k-Ew9p-3MnL7bZx4Wv2" # Replace with your actual API key or load from environment
|
|
model: "DeepSeek-R1" # Replace with the model you want to benchmark
|
|
|
|
requests: 10 # Total number of requests to send
|
|
concurrency: 5 # Number of concurrent workers
|
|
rateLimit: 0 # Requests per second (0 means no limit)
|
|
duration: 10s # Short duration for testing
|
|
|
|
promptTokens: 50 # Target number of tokens for the generated prompt (e.g., 50 or 1000)
|
|
maxTokens: 150 # Maximum tokens allowed in the response (if applicable)
|
|
streaming: true # Enable streaming responses
|
|
|
|
timeout: 30s # Timeout for each individual request
|
|
|
|
outputReport: "benchmark_report.html" # Filename for the generated HTML report
|
|
|
|
client: "fasthttp" # Choose the HTTP client: "fasthttp" (default) or "nethttp"
|
|
|
|
headers: # Optional: Custom HTTP headers
|
|
# Authorization: "Bearer YOUR_OTHER_TOKEN" # Example
|
|
# Custom-Header: "Value"
|
|
|
|
payload: # Optional: Custom fields to add/override in the request payload
|
|
# temperature: 0.7
|
|
# top_p: 0.9
|