LLM 1
Preview
Runtime: simulated vLLM
In workbench
Self-hosted Inference
Model Workbench
Preset credit$28.64
Interactive local preview with preset models, responses, and metrics. It never connects to the inference Engine or starts cloud compute.
Interactive previewPreset dataEngine not connected
Preview mode
All model names, answers, runtime states, credit, and performance values on this page are fixed demonstration data. Model simulations stay in this browser and never contact the AI Engine. This page checks live-demo eligibility once so the request button can show its current availability.
LLM 1 is open with preset performance metrics.
Model Registry
Preset modelsSimulated model states stay in this browser. Starting a model only demonstrates the interface flow.
Evaluation Run
Preset responsesRepeatable model evaluation
Run summary
3 parallel tasks · 8 total turns · each task runs serially · fixed preview answers
Progress
0/8
Elapsed
—
Task execution