NVIDIA/NeMo-Agent-Toolkit v1.5.0

v1.5.0

Repository: NVIDIA/NeMo-Agent-Toolkit

Tag: v1.5.0

Published: 2026-03-12T20:04:29Z

Prerelease: no

Release notes:

🚀 Notable Features and Improvements

• **Dynamo Runtime Intelligence:** Automatically infer per-request latency sensitivity from agent profiles and apply runtime hints for cache control, load-aware routing, and priority-aware serving.
• **Agent Performance Primitives (APP):** Introduce framework-agnostic performance primitives that accelerate graph-based agent frameworks such as LangChain, CrewAI, and Agno with parallel execution, speculative branching, and node-level priority routing.
• **LangSmith Native Integration:** Observe end-to-end agent execution with native LangSmith tracing, run evaluation experiments, compare outcomes, and manage prompt versions across development and production workflows.
• **FastMCP Workflow Publishing:** Publish NeMo Agent Toolkit workflows as MCP servers using the FastMCP runtime to simplify MCP-native deployment and integration.

Migration notice: 1.5.0 includes packaging and compatibility refactors (meta-package restructure, eval/profiling package split, and import-path updates). See the Migration Guide.

🚨 Breaking Changes

• Dynamic Inference Headers with Prediction Trie Integration by @dnandakumar-nv in https://github.com/NVIDIA/NeMo-Agent-Toolkit/pull/1483
• improvement(packaging): Packaging Restructure for libraries by @willkill07 in https://github.com/NVIDIA/NeMo-Agent-Toolkit/pull/1512
• fix: Langchain agents should reuse runnable config by @willkill07 in https://github.com/NVIDIA/NeMo-Agent-Toolkit/pull/1604
• Refactor: Split eval/profiler into optional nvidia-nat-eval package by @AnuradhaKaruppiah in https://github.com/NVIDIA/NeMo-Agent-Toolkit/pull/1599
• improvement: nvext.agent_hints and nvext.cache_control clean up by @bbednarski9 in https://github.com/NVIDIA/NeMo-Agent-Toolkit/pull/1648

✨ New Features

• Add evaluator feedback to GA optimizer by @dnandakumar-nv in https://github.com/NVIDIA/NeMo-Agent-Toolkit/pull/1442
• Add similarity scores and threshold filtering to Redis semantic search by @thepatrickchin in https://github.com/NVIDIA/NeMo-Agent-Toolkit/pull/1434
• Add end-to-end custom metadata propagation for observability by @ericevans-nv in https://github.com/NVIDIA/NeMo-Agent-Toolkit/pull/1480
• Expose MCP session ID and add custom headers support for session correlation by @yczhang-nv in https://github.com/NVIDIA/NeMo-Agent-Toolkit/pull/1500
• Preserve workflow state across WebSocket reconnections by @ericevans-nv in https://github.com/NVIDIA/NeMo-Agent-Toolkit/pull/1541
• feat(ci-scripts): utility scripts for license updates and SBOM by @willkill07 in https://github.com/NVIDIA/NeMo-Agent-Toolkit/pull/1548
• Add example of control flow composition with router agent and sequential executor by @thepatrickchin in https://github.com/NVIDIA/NeMo-Agent-Toolkit/pull/1550
• Add support for publishing a NeMo Agent Toolkit workflow as a MCP server via FastMCP3 by @AnuradhaKaruppiah in https://github.com/NVIDIA/NeMo-Agent-Toolkit/pull/1539
• Add dataset store registration and support by @dnandakumar-nv in https://github.com/NVIDIA/NeMo-Agent-Toolkit/pull/1576
• Allow for prompts to be stored in/loaded from files by @pastorsj in https://github.com/NVIDIA/NeMo-Agent-Toolkit/pull/1567
• feat(observability): Cross-Workflow Observability by @willkill07 in https://github.com/NVIDIA/NeMo-Agent-Toolkit/pull/1598
• FastAPI Frontend Refactor with HTTP HITL and OAuth Support by @willkill07 in https://github.com/NVIDIA/NeMo-Agent-Toolkit/pull/1603
• Add support for LangSmith evaluators by @mpenn in https://github.com/NVIDIA/NeMo-Agent-Toolkit/pull/1592
• Add automatic latency sensitivity inference by @dnandakumar-nv in https://github.com/NVIDIA/NeMo-Agent-Toolkit/pull/1618
• RAG Library Mode integration by @ericevans-nv in https://github.com/NVIDIA/NeMo-Agent-Toolkit/pull/1440
• feat: Add HuggingFace Inference API and Embedder providers by @bledden in https://github.com/NVIDIA/NeMo-Agent-Toolkit/pull/1570
• Integrate LangSmith Observability with Evaluation and Optimization by @pastorsj in https://github.com/NVIDIA/NeMo-Agent-Toolkit/pull/1593
• Add experimental nvidia-nat-app Agent Performance Primitives subpackage by @mpenn in https://github.com/NVIDIA/NeMo-Agent-Toolkit/pull/1636
• Add Dynamo Example for Latency Sensitivity Assignment by @dnandakumar-nv in https://github.com/NVIDIA/NeMo-Agent-Toolkit/pull/1634
• Revert unintended changes from PR #1704, preserve nat-ui submodule fix by @ericevans-nv in https://github.com/NVIDIA/NeMo-Agent-Toolkit/pull/1710

🔧 Improvements

• Restore version 1.5 on develop after forward merge by @mnajafian-nv in https://github.com/NVIDIA/NeMo-Agent-Toolkit/pull/1324
• Forward-merge release/1.4 into develop (conflict resolution) by @mnajafian-nv in https://github.com/NVIDIA/NeMo-Agent-Toolkit/pull/1394
• Add OAuth2-Protected MCP Calculator Example by @AnuradhaKaruppiah in https://github.com/NVIDIA/NeMo-Agent-Toolkit/pull/1403
• Forward-merge release/1.4 into develop by @mnajafian-nv in https://github.com/NVIDIA/NeMo-Agent-Toolkit/pull/1453
• Merge release/1.4 into develop by @willkill07 in https://github.com/NVIDIA/NeMo-Agent-Toolkit/pull/1459
• Add use_native_tool_calling option to ReAct agent by @yczhang-nv in https://github.com/NVIDIA/NeMo-Agent-Toolkit/pull/1476
• Add raise_on_parsing_failure option to ReAct agent by @yczhang-nv in https://github.com/NVIDIA/NeMo-Agent-Toolkit/pull/1477
• Enable per-user workflow support in nat eval by @ericevans-nv in https://github.com/NVIDIA/NeMo-Agent-Toolkit/pull/1503
• feat: make tavily internet search tool configurable by @cdgamarose-nv in https://github.com/NVIDIA/NeMo-Agent-Toolkit/pull/1518
• Update nat-ui submodule to latest main by @ericevans-nv in https://github.com/NVIDIA/NeMo-Agent-Toolkit/pull/1551
• chore(pre-commit): update…