For AI agents: a documentation index is available at the root level at /llms.txt and /llms-full.txt. Append /llms.txt to any URL for a page-level index, or .md for the markdown version of any page.
  • About NVIDIA NeMo Relay
    • Overview
    • Architecture
    • Ecosystem
    • Concepts
    • Release Notes
  • Getting Started
    • Agent Runtime Primer
    • Prerequisites
    • Installation
    • Configuration / Setup
    • Quick Start
  • NVIDIA NeMo Relay CLI
    • About
    • Basic Usage
    • Claude Code
    • Codex
    • Cursor
    • Hermes Agent
  • Supported Integrations
    • About
    • OpenClaw Plugin Guide
    • LangChain Integration Guide
    • LangGraph Integration Guide
    • Deep Agents Integration Guide
  • Instrument Applications
    • About
    • Adding Scopes and Marks
    • Instrument a Tool Call
    • Instrument an LLM Call
    • Add Middleware
    • Code Examples
  • Observability Plugin
    • About
    • Configuration
    • Agent Trajectory Interchange Format (ATIF)
    • Agent Trajectory Observability Format (ATOF)
    • OpenTelemetry
    • OpenInference
  • Adaptive Plugin
    • About
    • Configuration
    • Adaptive Cache Governor (ACG)
    • Adaptive Hints
  • NeMo Guardrails Plugin
    • About
    • Configuration
  • Integrate into Frameworks
    • About
    • Adding Scopes
    • Wrap Tool Calls
    • Wrap LLM Calls
    • Handle Non-Serializable Data
    • Using Codecs
    • Provider Codecs
    • Provider Response Codecs
    • Code Examples
  • Build Plugins
    • About
    • Define a Plugin
    • Validate Plugin Configuration
    • Plugin Configuration Files
    • Register Plugin Behavior
    • Design Plugin Configuration
    • NeMo Guardrails Example Plugin
    • Code Examples
  • Contribute
    • About
    • Development Setup
    • Workflow and Reviews
    • Testing and Documentation
  • Reference
    • APIs
      • Python Library Reference
      • Node.js Library Reference
      • Rust Library Reference
        • nemo-relay
        • nemo-relay-adaptive
          • acg
          • acg_component
          • acg_learner
          • acg_profile
          • adaptive_hints_intercept
          • cache_diagnostics
          • config
          • context_helpers
          • drain
          • error
          • intercepts
          • learner
          • plugin_component
          • redis
          • storage
          • subscriber
          • tool_parallelism_learner
          • trie
          • types
          • AdaptiveRuntime
        • nemo-relay-ffi
    • Performance
  • Resources
    • Support and FAQs
    • Glossary
    • Troubleshooting Guide
    • Community
    • Legal
NVIDIANVIDIA
Developer-friendly docs for your API
Privacy Policy | Your Privacy Choices | Terms of Service | Accessibility | Corporate Policies | Product Security | Contact

Copyright © 2026, NVIDIA Corporation.

LogoLogo
On this page
  • NeMo Relay Adaptive
  • Re-exports
  • Modules
  • Structs
ReferenceAPIsRust Library Reference

nemo-relay-adaptive

||View as Markdown|
Previous

Macro editor_config

Next

Module acg

Generated from cargo doc --no-deps -p nemo-relay -p nemo-relay-adaptive -p nemo-relay-ffi.

NeMo Relay Adaptive

Adaptive config helpers and core-plugin integration for NeMo Relay. Adaptive behavior is enabled through the generic core plugin system.

This crate provides the adaptive runtime, persistence abstractions, learner implementations, and Adaptive Cache Governor (ACG) analysis types used to derive and apply runtime hints from observed NeMo Relay executions.

Re-exports

  • pub use config::AcgComponentConfig;
  • pub use config::AdaptiveConfig;
  • pub use config::AdaptiveHintsComponentConfig;
  • pub use config::BackendSpec;
  • pub use config::StateConfig;
  • pub use config::TelemetryComponentConfig;
  • pub use config::ToolParallelismComponentConfig;
  • pub use context_helpers::LATENCY_SENSITIVITY_POINTER;
  • pub use context_helpers::extract_scope_path;
  • pub use context_helpers::read_manual_latency_sensitivity;
  • pub use context_helpers::resolve_agent_id;
  • pub use context_helpers::resolve_shared_parent_scope_identity;
  • pub use context_helpers::set_latency_sensitivity;
  • pub use error::AdaptiveError;
  • pub use error::Result;
  • pub use redis::RedisBackend;
  • pub use storage::erased::AnyBackend;
  • pub use storage::memory::InMemoryBackend;
  • pub use storage::traits::StorageBackend;
  • pub use storage::traits::StorageBackendDyn;

Modules

  • acg: Canonical Adaptive Cache Governor (ACG) module surface exposed from the adaptive crate.
  • acg_component: Adaptive Cache Governor (ACG) request and execution intercept helpers for the adaptive runtime.
  • acg_learner: Adaptive Cache Governor (ACG) learner for the adaptive telemetry pipeline.
  • acg_profile: Derives stable Adaptive Cache Governor (ACG) profile keys from structured LLM requests.
  • adaptive_hints_intercept: AdaptiveHintsIntercept: opt-in LLM request intercept that injects AgentHints from HotCache trie.
  • cache_diagnostics: Runtime-local cache miss request facts and diagnostics tracking.
  • config: Canonical adaptive config and diagnostics types.
  • context_helpers: Context helpers for reading scope metadata on the intercept hot path.
  • drain: Background drain task for async telemetry processing.
  • error: Error types for the nemo-relay-adaptive crate.
  • intercepts: Intercept factories for the nemo-relay-adaptive crate, including Adaptive Cache Governor (ACG) intercepts.
  • learner: Learning primitives and built-in learner implementations. Adaptive learners that derive runtime hints from observed executions.
  • plugin_component: Core plugin integration for the adaptive runtime.
  • redis: Redis-backed StorageBackend implementation.
  • storage: Storage backends and backend traits for adaptive state persistence. Storage backends for adaptive runtime state and learned artifacts.
  • subscriber: Event subscriber factory and event-to-record mapping helpers.
  • tool_parallelism_learner: Learner that derives tool fan-out plans from observed runs. Learner that derives tool parallelism plans from observed runs.
  • trie: Prediction trie data structures for the online learning engine.
  • types: Serializable adaptive data models shared across runtime components. Serializable adaptive runtime data models.

Structs

  • AdaptiveRuntime: Hosted adaptive runtime that registers NeMo Relay plugin components.