⚡ LLM Load Tester
Idle
Test Configuration
API URL
OpenAI-compatible streaming chat completions endpoint.
API Key
(optional)
API Type
OpenAI-compatible
Anthropic-compatible
Determines auth header style for the models endpoint.
Models
Fetch Models
Clear
Click "Fetch Models" to load the list from
/v1/models
.
0
model(s) selected.
Concurrencies
Comma-separated list of concurrent users.
Max Tokens
Requests / User
Seed
Think Time Min (s)
Think Time Max (s)
Questions
(one per line)
Clear
Upload a
.txt
file (one question per line) or paste directly.
Or generate questions
Source
Built-in bank
LLM
Count
Seed
Model
Topic
(optional)
Generate
Start Test
Stop
Run History
Refresh
Clear
Run Progress
Waiting to start…
0%
0
Requests Done
0
Successes
0
Errors
0
Tokens / sec
0
Requests / sec
–
TTFT mean (ms)
0
Output Tokens
0s
Elapsed
Results
Results will appear here once a run completes.
Results detail
All users
Errors only
#
Question
Answer
Status
TTFT (ms)
Latency (ms)
Out tok
In tok
Error