Transport
HTTP status: `200`
Response size: `778 bytes`
Benchmark
LLM latency baseline on March 30, 2026
This page records a short public request against the OpenAI-compatible chat endpoint on `probqa.com`. The purpose is to show the live request path, token accounting, and end-to-end latency for a small completion, not to claim maximum throughput.
Request
Public API call used for the baseline
curl https://probqa.com/v1/chat/completions \
-H "Authorization: Bearer cgs_live_your_key" \
-H "Content-Type: application/json" \
-d '{
"model": "meta-llama/Llama-4-Scout-17B-16E-Instruct",
"messages": [
{"role": "system", "content": "Be concise and useful."},
{"role": "user", "content": "Give a one-sentence description of an online SAT solver."}
],
"max_tokens": 96
}'
Result
Measured output
Usage
Prompt tokens: `32`
Completion tokens: `38`
Total tokens: `70`
Preview
“An online SAT (Satisfiability) solver is a web-based tool that takes a Boolean satisfiability problem as input and returns a solution or determines that the problem is unsatisfiable.”
Interpretation
Why this matters to buyers
- The public API path is live and responds with OpenAI-style JSON.
- Token accounting is visible and usable for prepaid billing.
- This is a small-latency baseline you can compare against your own prompts.
- For production decisions, test with your own prompt mix and response-length targets.
Next Step
Use the same curl shape on your own workload
Trial credits let you test the exact OpenAI-compatible surface with your own prompts, your own system messages, and your own token budgets.