[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-80970":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":10,"language":11,"languages":10,"totalLinesOfCode":10,"stars":12,"forks":13,"watchers":14,"openIssues":14,"contributorsCount":15,"subscribersCount":15,"size":15,"stars1d":14,"stars7d":16,"stars30d":17,"stars90d":15,"forks30d":15,"starsTrendScore":18,"compositeScore":19,"rankGlobal":10,"rankLanguage":10,"license":20,"archived":21,"fork":21,"defaultBranch":22,"hasWiki":23,"hasPages":23,"topics":24,"createdAt":10,"pushedAt":10,"updatedAt":25,"readmeContent":26,"aiSummary":27,"trendingCount":15,"starSnapshotCount":15,"syncStatus":14,"lastSyncTime":28,"discoverSource":29},80970,"nemo-platform","NVIDIA-NeMo\u002Fnemo-platform","NVIDIA-NeMo","Make the agents you ship faster, more accurate, and safer.","https:\u002F\u002Fnvidia-nemo.github.io\u002Fnemo-platform\u002Fmain\u002F",null,"Python",43,4,2,0,8,12,6,2.1,"Apache License 2.0",false,"main",true,[],"2026-06-12 02:04:09","# NeMo Platform\n\n![NEMO Platform](docs\u002Fassets\u002Fnemo-wordmark.svg)\n\n\u003C!-- Once the repo is public, swap the CI badge back to the dynamic GitHub Actions one:\n     [![CI](https:\u002F\u002Fgithub.com\u002FNVIDIA-NeMo\u002Fnemo-platform\u002Factions\u002Fworkflows\u002Fci.yaml\u002Fbadge.svg)](https:\u002F\u002Fgithub.com\u002FNVIDIA-NeMo\u002Fnemo-platform\u002Factions\u002Fworkflows\u002Fci.yaml) -->\n[![CI](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCI-passing-brightgreen?style=flat-square&logo=githubactions&logoColor=white)](https:\u002F\u002Fgithub.com\u002FNVIDIA-NeMo\u002Fnemo-platform\u002Factions\u002Fworkflows\u002Fci.yaml)\n[![License](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Flicense-Apache_2.0-D22128?style=flat-square)](LICENSE)\n[![Python](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Fpython-3.11--3.13-3776AB?style=flat-square&logo=python&logoColor=white)](https:\u002F\u002Fwww.python.org\u002F)\n[![Docs](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Fdocs-nvidia--nemo.github.io-76B900?style=flat-square&logo=readthedocs&logoColor=white)](https:\u002F\u002Fnvidia-nemo.github.io\u002Fnemo-platform\u002F)\n\nMake the agents you ship faster, more accurate, and safer.\n\nNeMo Platform brings NVIDIA NeMo libraries together under one CLI, Python SDK, and web UI. Hardening, evaluation, and tuning for the agents you put in production.\n\n## Get started\n\n**Prerequisites:** Python 3.11-3.13, uv, and an API key for an inference provider (NVIDIA Build, OpenAI, Anthropic, Google Gemini, or a local Ollama instance). Node.js 22.18.x with `pnpm` only if you want the web UI.\n\n```bash\ngit clone https:\u002F\u002Fgithub.com\u002FNVIDIA-NeMo\u002Fnemo-platform.git\ncd nemo-platform\n\nmake bootstrap\nsource .venv\u002Fbin\u002Factivate\n\nnemo setup\n```\n\n`nemo setup` starts local services, registers your LLM provider, discovers available models, installs agent skills, and deploys a sample agent (see more below).\n\nSee **[skills\u002Fnemo-setup\u002FSKILL.md](skills\u002Fnemo-setup\u002FSKILL.md)** for the full setup playbook (local data dir, DB reset, manual service start, troubleshooting).\n\nVerify:\n\n```bash\nnemo services status\n```\n\nTo permanently reset the database state: `rm -rf ~\u002F.local\u002Fshare\u002Fnemo`.\n\n\u003Cdetails>\n\u003Csummary>Useful CLI commands once setup completes\u003C\u002Fsummary>\n\n```bash\nnemo --help                # All commands\nnemo models list           # Available models\nnemo chat \u003Cmodel-name>     # Chat directly with a model\nnemo services status       # Platform health\nnemo skills list           # Skills installed on the platform\n```\n\nEvery capability is also available via REST API. Model inference uses the model IDs returned from `nemo models list` and is available at:\n\n```text\nhttp:\u002F\u002Flocalhost:8080\u002Fapis\u002Finference-gateway\u002Fv2\u002Fworkspaces\u002Fdefault\u002Fopenai\u002F-\u002Fv1\u002Fchat\u002Fcompletions\n```\n\nTo run platform services in the foreground in a separate terminal (instead of the background process `nemo setup` starts):\n\n```bash\nnemo services run\n```\n\n\u003C\u002Fdetails>\n\n\u003Cdetails>\n\u003Csummary>Studio (web UI) bootstrap troubleshooting\u003C\u002Fsummary>\n\nIf `make bootstrap` reports that Studio asset bootstrap did not complete, the API still runs but the web UI is unavailable until the bundle is built. Install Node 22.18.x with `pnpm env use --global 22.18.0`, then run `make bootstrap-studio` from the repo root.\n\n\u003C\u002Fdetails>\n\n\u003Cdetails>\n\u003Csummary>Non-interactive setup (for agents, CI, or scripts)\u003C\u002Fsummary>\n\n```bash\nexport NVIDIA_API_KEY=nvapi...\nexport NEMO_DEFAULT_MODEL=nvidia-nemotron-3-super-120b-a12b\nnemo setup --auto --start-services --install-skills --deploy-agent\n```\n\n\u003C\u002Fdetails>\n\n## Use NeMo Platform from your coding agent\n\nAfter installation, launch your coding agent (Claude Code, Codex, Cursor, OpenCode, etc) from inside the `nemo-platform` directory. This is the primary way of interacting with the NeMo Platform.\n\nThings you can ask it to do, once the platform is running:\n\n- \"Scaffold an agent from this spec and deploy it.\"\n- \"Run an evaluation against my agent.\"\n- \"Add content-safety guardrails to my agent.\"\n- \"Help me optimize my agent.\"\n- \"Show me what's running on the platform.\"\n- \"Shut down NeMo cleanly.\"\n\n## What's here today\n\n- **Secure agents.** Guardrails (content safety, jailbreak detection, PII redaction), Auditor (red-teaming via garak), Anonymizer (PII handling for training data).\n- **Evaluate agents.** LLM-as-judge, deterministic, agentic, and RAG benchmarks. Harbor-backed eval suites for regression testing.\n- **Tune agents.** Skill optimization, prompt and hyperparameter tuning, Switchyard model routing.\n- **Build agents.** NVIDIA NeMo Agent Toolkit (NAT) for LangGraph-based agents. Shared infrastructure: Inference Gateway, Secrets, Files, Entity Store, Jobs.\n- **Generate synthetic data.** Generate synthetic data for training or evaluation purposes using Data Designer.\n- **NeMo Studio (alpha).** Installed automatically with the platform. Browser UI for chat, monitoring, and reviewing optimization suggestions. Studio's agent-focused features are still a work in progress; the CLI is the primary surface today.\n\n## What's coming soon\n\n- Fine-tuning\n- Safe Synthesizer (synthetic data with privacy guarantees)\n- Broader agent framework support. Today NeMo Platform optimizes LangGraph agents wrapped in NAT. If your agent is in another framework, you need to write the NAT wrapper.\n\n## Skills\n\n`nemo setup` detects Claude Code, Cursor, Codex, and OpenCode and installs NeMo skills into your agent of choice, either into the local directory or globally. Platform-level skills live under `packages\u002Fnemo_platform_ext\u002Fsrc\u002Fnemo_platform_ext\u002Fskills\u002F` and ship with the `nemo-platform` package; plugin-owned skills live under `plugins\u002F\u003Cplugin>\u002Fsrc\u002F\u003Cplugin>\u002Fskills\u002F`.\n\nTo install or refresh skills:\n\n```bash\nnemo skills install --agent claude\nnemo skills install --agent claude --skill nemo-build-agent --skill nemo-status\n```\n\n## Try the demo agent\n\n`nemo setup --deploy-agent` deploys a demo calculator agent you can use to\nexplore the platform's evaluate \u002F optimize loop.\n\n```bash\nnemo agents invoke --agent calculator-agent --input \"what is 12 * 8?\"\n```\n\nThe calculator-agent package is installed automatically (`plugins\u002Fnemo-agents\u002Fexamples\u002Fcalculator-agent\u002F`).\n\n\u003Cdetails>\n\u003Csummary>Deploy it manually\u003C\u002Fsummary>\n```bash\nnemo agents create --name calculator-agent \\\n  --agent-config plugins\u002Fnemo-agents\u002Fexamples\u002Fcalculator-agent\u002Fsrc\u002Fcalculator_agent\u002Fcalculator-agent.yml\nnemo agents deploy --agent calculator-agent\nnemo agents deployments wait --agent calculator-agent\n```\n\u003C\u002Fdetails>\n\n\u003Cdetails>\n\u003Csummary>Evaluate the agent\u003C\u002Fsummary>\n```bash\nnemo agents evaluate run \\\n  --eval-config plugins\u002Fnemo-agents\u002Fexamples\u002Fcalculator-agent\u002Fsrc\u002Fcalculator_agent\u002Fcalculator-eval.yml \\\n  --agent calculator-agent\n```\n\u003C\u002Fdetails>\n\n\u003Cdetails>\n\u003Csummary>Optimize the agent\u003C\u002Fsummary>\n```bash\nnemo agents optimize run \\\n  --optimize-config plugins\u002Fnemo-agents\u002Fexamples\u002Fcalculator-agent\u002Fsrc\u002Fcalculator_agent\u002Fcalculator-optimize.yml \\\n  --agent calculator-agent\n```\n\u003C\u002Fdetails>\n\nThe demo agent uses `${NEMO_DEFAULT_MODEL}` for both execution and the judge LLM. To select different models for either\u002Fboth, update the yaml config files.\n\n## Documentation\n\nFull documentation: [NeMo Platform docs](https:\u002F\u002Fnvidia-nemo.github.io\u002Fnemo-platform\u002F).\n\n- [Setup](docs\u002Fget-started\u002Fsetup.md): installation, providers, SDK.\n- [CLI reference](docs\u002Fcli\u002Findex.md): all commands.\n- [API reference](docs\u002Fapi\u002Findex.md): REST endpoints.\n\n## Development\n\nSee [CONTRIBUTING.md](CONTRIBUTING.md) for development workflow.\nSee [TESTING.md](TESTING.md) for testing strategy.\n\n## License\n\nNeMo Platform is licensed under the Apache License 2.0. Third-party open-source dependencies have their own licenses; review them before use.\n","NeMo Platform 是一个用于加速、提高准确性和增强安全性的代理开发平台。它通过统一的命令行界面（CLI）、Python SDK 和网页用户界面（UI），将NVIDIA NeMo库整合在一起，提供了一套完整的工具链来支持代理程序的加固、评估和调优过程。该平台支持多种大语言模型提供商，并允许用户轻松地管理和部署基于这些模型的应用程序。适用于需要快速构建和优化AI驱动应用的场景，如客户服务聊天机器人、自动化助手等。","2026-06-11 04:03:02","CREATED_QUERY"]