> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.nvidia.com/aiperf/dev/reference/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.nvidia.com/aiperf/_mcp/server.

# Reference

## Docs

- [Command Line Options](https://docs.nvidia.com/aiperf/dev/reference/command-line-options.md)
- [Environment Variables](https://docs.nvidia.com/aiperf/dev/reference/environment-variables.md)
- [Metrics Reference](https://docs.nvidia.com/aiperf/dev/reference/ai-perf-metrics-reference.md)
- [GPU Telemetry & Power-Efficiency Dataflow](https://docs.nvidia.com/aiperf/dev/reference/gpu-telemetry-power-efficiency-dataflow.md)
- [Benchmark Datasets](https://docs.nvidia.com/aiperf/dev/reference/benchmark-datasets.md)
- [Pre-Flight Tokenizer Auto Detection](https://docs.nvidia.com/aiperf/dev/reference/pre-flight-tokenizer-auto-detection.md)
- [Input Sequence Length (ISL) Tokenization](https://docs.nvidia.com/aiperf/dev/reference/input-sequence-length-isl-tokenization.md)
- [Prompt Corpus Selection](https://docs.nvidia.com/aiperf/dev/reference/prompt-corpus-selection.md)
- [Validating ISL/OSL Distributions with the Mock Server](https://docs.nvidia.com/aiperf/dev/reference/validating-isl-osl-distributions-with-the-mock-server.md)
- [ISL Budget Compensation Derivation](https://docs.nvidia.com/aiperf/dev/reference/isl-budget-compensation-derivation.md)
- [Conversation Context Mode](https://docs.nvidia.com/aiperf/dev/reference/conversation-context-mode.md)
- [List-Metric Aggregation](https://docs.nvidia.com/aiperf/dev/reference/list-metric-aggregation.md)
- [Effective vs Active Metrics](https://docs.nvidia.com/aiperf/dev/reference/effective-vs-active-metrics.md)
- [Vendor Usage Field Reference](https://docs.nvidia.com/aiperf/dev/reference/vendor-usage-field-reference.md)
- [Per-Request Speculative-Decoding Acceptance](https://docs.nvidia.com/aiperf/dev/reference/per-request-speculative-decoding-acceptance.md)
- [JSON Export Schema](https://docs.nvidia.com/aiperf/dev/reference/json-export-schema.md)
- [HTTP API Endpoints](https://docs.nvidia.com/aiperf/dev/reference/http-api-endpoints.md)
- [Cache-Bust Targets](https://docs.nvidia.com/aiperf/dev/reference/cache-bust-targets.md)
- [YAML Config Roadmap](https://docs.nvidia.com/aiperf/dev/reference/yaml-config-roadmap.md)
- [Mocking a ROCm Environment](https://docs.nvidia.com/aiperf/dev/reference/mocking-a-ro-cm-environment.md)