**`--log-stats` is required to populate metric data.** Without the flag, the endpoint still returns `200 OK` but the upstream `vllm:*` wrap is not registered at all, and the 15 `vllm_omni:*` families are registered as placeholders with no data written to them. This keeps the runtime cost essentially zero for deployments that don't need monitoring. With the flag, all ~80 families populate.
Emitted at request finalize, except for `audio_ttfp_s` (the first audio packet/frame from streaming Chat or Speech APIs) and `audio_underrun_s` / `audio_continuity_ok_total` (Chat or Speech streaming finalize, after the chunk stream is exhausted). Non-streaming Speech latency is represented by `e2e_request_latency_s`, not TTFP. All audio metrics carry `{model_name, stage, replica}` plus the listed extra label.
The continuity math comes from `vllm_omni/benchmarks/audio_continuity.py::compute_continuity_stats` so the server-side observation aligns with the bench-side definition.
For Speech, continuity applies to raw audio, SSE, and WebSocket streaming. WAV headers and empty chunks are excluded; non-streaming Speech requests do not emit continuity samples.
Continuity rate for the default 100 ms threshold can be derived from the successful-continuity counter and the underrun Histogram's request count:
Per-request timing breakdowns for diffusion (image/video) stages. Emitted at request finalize when `stage_metrics.diffusion_metrics` is present. All carry `{model_name, stage, replica}`.
| `vllm_omni:diffusion_exec_s` | Histogram | DiT forward pass execution time per request in seconds |
| `vllm_omni:diffusion_exec_per_step_s` | Histogram | DiT forward pass execution time per denoising step in seconds (`exec_s / num_inference_steps`) |
| `vllm_omni:diffusion_preprocess_s` | Histogram | Diffusion input preprocessing time per request in seconds |
| `vllm_omni:diffusion_postprocess_s` | Histogram | Diffusion output postprocessing (VAE decode) time per request in seconds |
`diffusion_exec_s` uses `SECONDS_BUCKETS` (50 ms → 300 s); the other three use `SECONDS_FAST_BUCKETS` (1 ms → 60 s). `diffusion_exec_per_step_s` is skipped when `num_inference_steps` is not set.
Per-physical-transfer histograms tracking the data hop between adjacent stages. Labels `{model_name, from_stage, from_replica, to_stage, to_replica}` let dashboards attribute latency to specific replica edges. `from_replica` / `to_replica` are resolved from the orchestrator's sticky-routing binding (`stage_pool.get_bound_replica_id(request_id)`), so no extra plumbing through `TransferEdgeStats` is needed.
When the pipeline includes an LLM stage, the upstream vLLM engine exposes its full set of ~37 metric families under the `vllm:` prefix.
vLLM-Omni wraps the upstream `vllm.v1.metrics.loggers.PrometheusStatLogger` with `OmniPrometheusStatLogger` so that the original `engine` single label is reshaped into `stage` + `replica`. Every `vllm:*` family — TTFT, ITL, TPOT, e2e latency, KV cache usage, scheduler running/waiting, request success counts, etc. — therefore gains per-`(stage, replica)` visibility automatically. No omni-side duplicate is needed for the text path.
| `vllm:` MFU metrics | With `--log-stats --enable-mfu-metrics` |
## Naming Convention
- All time-bearing metrics use the `_s` suffix (values in seconds). Buckets are `SECONDS_BUCKETS` for e2e / generation-style values and `SECONDS_FAST_BUCKETS` (1 ms → 60 s) for the fine-grained transfer and audio-underrun values.
- Counters use the `_total` suffix (auto-appended by `prometheus_client`).
- All omni-specific families are prefixed `vllm_omni:`. The upstream `unregister_vllm_metrics()` function is monkey-patched (see `vllm_omni/patch.py`) to a scoped version that still strips upstream `vllm:*` collectors so multi-engine init within one process does not crash on duplicate registration, but preserves anything prefixed `vllm_omni:`.
- Text and audio first-output use distinct families (`vllm:time_to_first_token_seconds` reused from upstream for text; `vllm_omni:audio_ttfp_s` for audio) rather than a single metric with a `modality` label.
