Skip to content

instrumented_engine

diapason.telemetry.instrumented_engine

Transparent telemetry wrapper for inference engines.

InstrumentedEngine

Bases: InferenceEngine

Transparent wrapper that records telemetry around engine calls.

Agents call engine.generate() normally -- they don't know about telemetry. The wrapper publishes INFERENCE_START, INFERENCE_END, and TELEMETRY_RECORD events on the bus.

If an energy_monitor is provided (new multi-vendor :class:~diapason.telemetry.energy_monitor.EnergyMonitor), it is preferred over the legacy gpu_monitor for energy measurement.

generate(messages, *, model='', temperature=0.7, max_tokens=1024, **kwargs)

Generate with telemetry recording.

stream(messages, *, model='', temperature=0.7, max_tokens=1024, **kwargs) async

Stream with per-token timing and full telemetry recording.

stream_full(messages, *, model, temperature=0.7, max_tokens=1024, **kwargs) async

Delegate to inner engine's stream_full for tool-call support.