[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-71270":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":15,"contributorsCount":16,"subscribersCount":16,"size":16,"stars1d":17,"stars7d":18,"stars30d":19,"stars90d":16,"forks30d":16,"starsTrendScore":20,"compositeScore":21,"rankGlobal":10,"rankLanguage":10,"license":22,"archived":23,"fork":23,"defaultBranch":24,"hasWiki":25,"hasPages":23,"topics":26,"createdAt":10,"pushedAt":10,"updatedAt":41,"readmeContent":42,"aiSummary":43,"trendingCount":16,"starSnapshotCount":16,"syncStatus":44,"lastSyncTime":45,"discoverSource":46},71270,"gateway","Portkey-AI\u002Fgateway","Portkey-AI","A blazing fast AI Gateway with integrated guardrails. Route to 1,600+ LLMs, 50+ AI Guardrails with 1 fast & friendly API.","https:\u002F\u002Fportkey.ai\u002Ffeatures\u002Fai-gateway",null,"TypeScript",12026,1121,57,86,0,43,105,345,129,44.15,"MIT License",false,"main",true,[27,5,28,29,30,31,32,33,34,35,36,37,38,39,40],"ai-gateway","generative-ai","hacktoberfest","langchain","llm","llm-gateway","llmops","llms","mcp","mcp-client","mcp-gateway","mcp-servers","model-router","openai","2026-06-12 02:02:50","\n\u003Cp align=\"right\">\n   \u003Cstrong>English\u003C\u002Fstrong> | \u003Ca href=\".\u002F.github\u002FREADME.cn.md\">中文\u003C\u002Fa> | \u003Ca href=\".\u002F.github\u002FREADME.jp.md\">日本語\u003C\u002Fa>\n\u003C\u002Fp>\n\n> [!IMPORTANT]\n> :rocket: Gateway 2.0 (Pre-Release) Portkey's core enterprise gateway is merging into open-source with our 2.0 release. You can try the pre-release branch [here](https:\u002F\u002Fgithub.com\u002Fportkey-ai\u002Fgateway\u002Ftree\u002F2.0.0).\n> Read more about what's next for Portkey in our [**Series A announcement**](https:\u002F\u002Fportkey.wiki\u002Frohit-a).\n\n\n\u003Cdiv align=\"center\">\n\n🆕 **[Portkey Models](https:\u002F\u002Fgithub.com\u002FPortkey-AI\u002Fmodels)** - Open-source LLM pricing for 2,300+ models across 40+ providers. [Explore →](https:\u002F\u002Fportkey.ai\u002Fmodels)\n\n\n# AI Gateway\n#### Route to 250+ LLMs with 1 fast & friendly API\n\n\u003Cimg src=\"https:\u002F\u002Fcfassets.portkey.ai\u002Fsdk.gif\" width=\"550px\" alt=\"Portkey AI Gateway Demo showing LLM routing capabilities\" style=\"margin-left:-35px\">\n\n[Docs](https:\u002F\u002Fportkey.wiki\u002Fgh-1) | [Enterprise](https:\u002F\u002Fportkey.wiki\u002Fgh-2) | [Hosted Gateway](https:\u002F\u002Fportkey.wiki\u002Fgh-3) | [Changelog](https:\u002F\u002Fportkey.wiki\u002Fgh-4) | [API Reference](https:\u002F\u002Fportkey.wiki\u002Fgh-5)\n\n\n[![License](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Flicense\u002FIleriayo\u002Fmarkdown-badges)](.\u002FLICENSE)\n[![Discord](https:\u002F\u002Fimg.shields.io\u002Fdiscord\u002F1143393887742861333)](https:\u002F\u002Fportkey.wiki\u002Fgh-6)\n[![Twitter](https:\u002F\u002Fimg.shields.io\u002Ftwitter\u002Furl\u002Fhttps\u002Ftwitter\u002Ffollow\u002Fportkeyai?style=social&label=Follow%20%40PortkeyAI)](https:\u002F\u002Fportkey.wiki\u002Fgh-7)\n[![npm version](https:\u002F\u002Fbadge.fury.io\u002Fjs\u002F%40portkey-ai%2Fgateway.svg)](https:\u002F\u002Fportkey.wiki\u002Fgh-8)\n[![Better Stack Badge](https:\u002F\u002Fuptime.betterstack.com\u002Fstatus-badges\u002Fv1\u002Fmonitor\u002Fq94g.svg)](https:\u002F\u002Fportkey.wiki\u002Fgh-9)\n\n\u003Ca href=\"https:\u002F\u002Fus-east-1.console.aws.amazon.com\u002Fcloudformation\u002Fhome?region=us-east-1#\u002Fstacks\u002Fquickcreate?stackName=portkey-gateway&templateURL=https:\u002F\u002Fportkey-gateway-ec2-quicklaunch.s3.us-east-1.amazonaws.com\u002Fportkey-gateway-ec2-quicklaunch.template.yaml\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDeploy_to_EC2-232F3E?style=for-the-badge&logo=amazonwebservices&logoColor=white\" alt=\"Deploy to AWS EC2\" width=\"105\"\u002F>\u003C\u002Fa> [![Ask DeepWiki](https:\u002F\u002Fdeepwiki.com\u002Fbadge.svg)](https:\u002F\u002Fdeepwiki.com\u002FPortkey-AI\u002Fgateway)\n\u003C\u002Fdiv>\n\n\u003Cbr\u002F>\n\nThe [**AI Gateway**](https:\u002F\u002Fportkey.wiki\u002Fgh-10) is designed for fast, reliable & secure routing to 1600+ language, vision, audio, and image models. It is a lightweight, open-source, and enterprise-ready solution that allows you to integrate with any language model in under 2 minutes.\n\n- [x] **Blazing fast** (\u003C1ms latency) with a tiny footprint (122kb)\n- [x] **Battle tested**, with over 10B tokens processed everyday\n- [x] **Enterprise-ready** with enhanced security, scale, and custom deployments\n\n\u003Cbr>\n\n#### What can you do with the AI Gateway?\n- Integrate with any LLM in under 2 minutes - [Quickstart](#quickstart-2-mins)\n- Prevent downtimes through **[automatic retries](https:\u002F\u002Fportkey.wiki\u002Fgh-11)** and **[fallbacks](https:\u002F\u002Fportkey.wiki\u002Fgh-12)**\n- Scale AI apps with **[load balancing](https:\u002F\u002Fportkey.wiki\u002Fgh-13)** and **[conditional routing](https:\u002F\u002Fportkey.wiki\u002Fgh-14)**\n- Protect your AI deployments with **[guardrails](https:\u002F\u002Fportkey.wiki\u002Fgh-15)**\n- Go beyond text with **[multi-modal capabilities](https:\u002F\u002Fportkey.wiki\u002Fgh-16)**\n- Explore **[agentic workflow](https:\u002F\u002Fportkey.wiki\u002Fgh-17)** integrations\n- Manage MCP servers with enterprise auth & observability using **[MCP Gateway](https:\u002F\u002Fportkey.ai\u002Fdocs\u002Fproduct\u002Fmcp-gateway)**\n\n\u003Cbr>\u003Cbr>\n\n> [!TIP]\n> Starring this repo helps more developers discover the AI Gateway 🙏🏻\n>\n> ![star-2](https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F53597dce-6333-4ecc-a154-eb05532954e4)\n> \n\u003Cbr>\n\n\n\u003Cbr>\n\n## Quickstart (2 mins)\n\n### 1. Setup your AI Gateway\n\n```bash\n# Run the gateway locally (needs Node.js and npm)\nnpx @portkey-ai\u002Fgateway\n```\n> The Gateway is running on `http:\u002F\u002Flocalhost:8787\u002Fv1`\n> \n> The Gateway Console is running on `http:\u002F\u002Flocalhost:8787\u002Fpublic\u002F`\n\n\u003Csup>\nDeployment guides:\n  \u003Ca href=\"https:\u002F\u002Fportkey.wiki\u002Fgh-18\">\u003Cimg height=\"12\" width=\"12\" src=\"https:\u002F\u002Fcfassets.portkey.ai\u002Flogo\u002Fdew-color.svg\" \u002F> Portkey Cloud (Recommended)\u003C\u002Fa>\n  \u003Ca href=\".\u002Fdocs\u002Finstallation-deployments.md#docker\">\u003Cimg height=\"12\" width=\"12\" src=\"https:\u002F\u002Fcdn.simpleicons.org\u002Fdocker\u002F3776AB\" \u002F> Docker\u003C\u002Fa>\n  \u003Ca href=\".\u002Fdocs\u002Finstallation-deployments.md#nodejs-server\">\u003Cimg height=\"12\" width=\"12\" src=\"https:\u002F\u002Fcdn.simpleicons.org\u002Fnode.js\u002F3776AB\" \u002F> Node.js\u003C\u002Fa>\n  \u003Ca href=\".\u002Fdocs\u002Finstallation-deployments.md#cloudflare-workers\">\u003Cimg height=\"12\" width=\"12\" src=\"https:\u002F\u002Fcdn.simpleicons.org\u002Fcloudflare\u002F3776AB\" \u002F> Cloudflare\u003C\u002Fa>\n  \u003Ca href=\".\u002Fdocs\u002Finstallation-deployments.md#replit\">\u003Cimg height=\"12\" width=\"12\" src=\"https:\u002F\u002Fcdn.simpleicons.org\u002Freplit\u002F3776AB\" \u002F> Replit\u003C\u002Fa>\n  \u003Ca href=\".\u002Fdocs\u002Finstallation-deployments.md\"> Others...\u003C\u002Fa>\n\n\u003C\u002Fsup>\n\n### 2. Make your first request\n\n\u003C!-- \u003Cdetails open>\n\u003Csummary>Python Example\u003C\u002Fsummary> -->\n```python\n# pip install -qU portkey-ai\n\nfrom portkey_ai import Portkey\n\n# OpenAI compatible client\nclient = Portkey(\n    provider=\"openai\", # or 'anthropic', 'bedrock', 'groq', etc\n    Authorization=\"sk-***\" # the provider API key\n)\n\n# Make a request through your AI Gateway\nclient.chat.completions.create(\n    messages=[{\"role\": \"user\", \"content\": \"What's the weather like?\"}],\n    model=\"gpt-4o-mini\"\n)\n```\n\n\n\n\u003Csup>Supported Libraries:\n  [\u003Cimg height=\"12\" width=\"12\" src=\"https:\u002F\u002Fcdn.simpleicons.org\u002Fjavascript\u002F3776AB\" \u002F> JS](https:\u002F\u002Fportkey.wiki\u002Fgh-19)\n  [\u003Cimg height=\"12\" width=\"12\" src=\"https:\u002F\u002Fcdn.simpleicons.org\u002Fpython\u002F3776AB\" \u002F> Python](https:\u002F\u002Fportkey.wiki\u002Fgh-20)\n  [\u003Cimg height=\"12\" width=\"12\" src=\"https:\u002F\u002Fcdn.simpleicons.org\u002Fgnubash\u002F3776AB\" \u002F> REST](https:\u002F\u002Fportkey.sh\u002Fgh-84)\n  [\u003Cimg height=\"12\" width=\"12\" src=\"https:\u002F\u002Fcdn.simpleicons.org\u002Fopenai\u002F3776AB\" \u002F> OpenAI SDKs](https:\u002F\u002Fportkey.wiki\u002Fgh-21)\n  [\u003Cimg height=\"12\" width=\"12\" src=\"https:\u002F\u002Fcdn.simpleicons.org\u002Flangchain\u002F3776AB\" \u002F> Langchain](https:\u002F\u002Fportkey.wiki\u002Fgh-22)\n  [LlamaIndex](https:\u002F\u002Fportkey.wiki\u002Fgh-23)\n  [Autogen](https:\u002F\u002Fportkey.wiki\u002Fgh-24)\n  [CrewAI](https:\u002F\u002Fportkey.wiki\u002Fgh-25)\n  [More..](https:\u002F\u002Fportkey.wiki\u002Fgh-26)\n\u003C\u002Fsup>\n\nOn the Gateway Console (`http:\u002F\u002Flocalhost:8787\u002Fpublic\u002F`) you can see all of your local logs in one place.\n\n\u003Cimg src=\"https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F362bc916-0fc9-43f1-a39e-4bd71aac4a3a\" width=\"400\" \u002F>\n\n\n### 3. Routing & Guardrails\n`Configs` in the LLM gateway allow you to create routing rules, add reliability and setup guardrails.\n```python\nconfig = {\n  \"retry\": {\"attempts\": 5},\n\n  \"output_guardrails\": [{\n    \"default.contains\": {\"operator\": \"none\", \"words\": [\"Apple\"]},\n    \"deny\": True\n  }]\n}\n\n# Attach the config to the client\nclient = client.with_options(config=config)\n\nclient.chat.completions.create(\n    model=\"gpt-4o-mini\",\n    messages=[{\"role\": \"user\", \"content\": \"Reply randomly with Apple or Bat\"}]\n)\n\n# This would always response with \"Bat\" as the guardrail denies all replies containing \"Apple\". The retry config would retry 5 times before giving up.\n```\n\u003Cdiv align=\"center\">\n\u003Cimg src=\"https:\u002F\u002Fportkey.ai\u002Fblog\u002Fcontent\u002Fimages\u002Fsize\u002Fw1600\u002F2024\u002F11\u002Fimage-15.png\" width=600 title=\"Request flow through Portkey's AI gateway with retries and guardrails\" alt=\"Request flow through Portkey's AI gateway with retries and guardrails\"\u002F>\n\u003C\u002Fdiv>\n\nYou can do a lot more stuff with configs in your AI gateway. [Jump to examples  →](https:\u002F\u002Fportkey.wiki\u002Fgh-27)\n\n\u003Cbr\u002F>\n\n### Enterprise Version (Private deployments)\n\n\u003Csup>\n\n[\u003Cimg height=\"12\" width=\"12\" src=\"https:\u002F\u002Fcfassets.portkey.ai\u002Famazon-logo.svg\" \u002F> AWS](https:\u002F\u002Fportkey.wiki\u002Fgh-28)\n  [\u003Cimg height=\"12\" width=\"12\" src=\"https:\u002F\u002Fcfassets.portkey.ai\u002Fazure-logo.svg\" \u002F> Azure](https:\u002F\u002Fportkey.wiki\u002Fgh-29)\n  [\u003Cimg height=\"12\" width=\"12\" src=\"https:\u002F\u002Fcdn.simpleicons.org\u002Fgooglecloud\u002F3776AB\" \u002F> GCP](https:\u002F\u002Fportkey.wiki\u002Fgh-30)\n  [\u003Cimg height=\"12\" width=\"12\" src=\"https:\u002F\u002Fcdn.simpleicons.org\u002Fredhatopenshift\u002F3776AB\" \u002F> OpenShift](https:\u002F\u002Fportkey.wiki\u002Fgh-31)\n  [\u003Cimg height=\"12\" width=\"12\" src=\"https:\u002F\u002Fcdn.simpleicons.org\u002Fkubernetes\u002F3776AB\" \u002F> Kubernetes](https:\u002F\u002Fportkey.wiki\u002Fgh-85)\n\n\u003C\u002Fsup>\n\nThe LLM Gateway's [enterprise version](https:\u002F\u002Fportkey.wiki\u002Fgh-86) offers advanced capabilities for **org management**, **governance**, **security** and [more](https:\u002F\u002Fportkey.wiki\u002Fgh-87) out of the box. [View Feature Comparison →](https:\u002F\u002Fportkey.wiki\u002Fgh-32)\n\nThe enterprise deployment architecture for supported platforms is available here - [**Enterprise Private Cloud Deployments**](https:\u002F\u002Fportkey.ai\u002Fdocs\u002Fself-hosting\u002Fhybrid-deployments\u002Farchitecture)\n\n\u003Ca href=\"https:\u002F\u002Fportkey.sh\u002Fdemo-13\">\u003Cimg src=\"https:\u002F\u002Fportkey.ai\u002Fblog\u002Fcontent\u002Fimages\u002F2024\u002F08\u002FGet-API-Key--5-.png\" height=50 alt=\"Book an enterprise AI gateway demo\" \u002F>\u003C\u002Fa>\u003Cbr\u002F>\n\n\u003Cbr>\n\n## MCP Gateway\n\n[MCP Gateway](https:\u002F\u002Fportkey.ai\u002Fdocs\u002Fproduct\u002Fmcp-gateway) provides a centralized control plane for managing MCP (Model Context Protocol) servers across your organization.\n\n- **Authentication** — Single auth layer at the gateway. Users authenticate once; your MCP servers receive verified requests\n- **Access Control** — Control which teams and users can access which servers and tools. Revoke access instantly\n- **Observability** — Every tool call logged with full context: who called what, parameters, response, latency\n- **Identity Forwarding** — Forward user identity (email, team, roles) to MCP servers automatically\n\nWorks with Claude Desktop, Cursor, VS Code, and any MCP-compatible client. [Get started →](https:\u002F\u002Fportkey.ai\u002Fdocs\u002Fproduct\u002Fmcp-gateway\u002Fquickstart)\n\n\u003Cbr>\n\n## Core Features\n### Reliable Routing\n- \u003Ca href=\"https:\u002F\u002Fportkey.wiki\u002Fgh-37\">**Fallbacks**\u003C\u002Fa>: Fallback to another provider or model on failed requests using the LLM gateway. You can specify the errors on which to trigger the fallback. Improves reliability of your application.\n- \u003Ca href=\"https:\u002F\u002Fportkey.wiki\u002Fgh-38\">**Automatic Retries**\u003C\u002Fa>: Automatically retry failed requests up to 5 times. An exponential backoff strategy spaces out retry attempts to prevent network overload.\n- \u003Ca href=\"https:\u002F\u002Fportkey.wiki\u002Fgh-39\">**Load Balancing**\u003C\u002Fa>: Distribute LLM requests across multiple API keys or AI providers with weights to ensure high availability and optimal performance.\n- \u003Ca href=\"https:\u002F\u002Fportkey.wiki\u002Fgh-40\">**Request Timeouts**\u003C\u002Fa>: Manage unruly LLMs & latencies by setting up granular request timeouts, allowing automatic termination of requests that exceed a specified duration.\n- \u003Ca href=\"https:\u002F\u002Fportkey.wiki\u002Fgh-41\">**Multi-modal LLM Gateway**\u003C\u002Fa>: Call vision, audio (text-to-speech & speech-to-text), and image generation models from multiple providers  — all using the familiar OpenAI signature\n- \u003Ca href=\"https:\u002F\u002Fportkey.wiki\u002Fgh-42\">**Realtime APIs**\u003C\u002Fa>: Call realtime APIs launched by OpenAI through the integrate websockets server.\n\n### Security & Accuracy\n- \u003Ca href=\"https:\u002F\u002Fportkey.wiki\u002Fgh-88\">**Guardrails**\u003C\u002Fa>: Verify your LLM inputs and outputs to adhere to your specified checks. Choose from the 40+ pre-built guardrails to ensure compliance with security and accuracy standards. You can \u003Ca href=\"https:\u002F\u002Fportkey.wiki\u002Fgh-43\">bring your own guardrails\u003C\u002Fa> or choose from our \u003Ca href=\"https:\u002F\u002Fportkey.wiki\u002Fgh-44\">many partners\u003C\u002Fa>.\n- [**Secure Key Management**](https:\u002F\u002Fportkey.wiki\u002Fgh-45): Use your own keys or generate virtual keys on the fly.\n- [**Role-based access control**](https:\u002F\u002Fportkey.wiki\u002Fgh-46): Granular access control for your users, workspaces and API keys.\n- \u003Ca href=\"https:\u002F\u002Fportkey.wiki\u002Fgh-47\">**Compliance & Data Privacy**\u003C\u002Fa>: The AI gateway is SOC2, HIPAA, GDPR, and CCPA compliant.\n\n### Cost Management\n- [**Smart caching**](https:\u002F\u002Fportkey.wiki\u002Fgh-48): Cache responses from LLMs to reduce costs and improve latency. Supports simple and semantic* caching.\n- [**Usage analytics**](https:\u002F\u002Fportkey.wiki\u002Fgh-49): Monitor and analyze your AI and LLM usage, including request volume, latency, costs and error rates.\n- [**Provider optimization***](https:\u002F\u002Fportkey.wiki\u002Fgh-89): Automatically switch to the most cost-effective provider based on usage patterns and pricing models.\n\n### Collaboration & Workflows\n- \u003Ca href=\"https:\u002F\u002Fportkey.ai\u002Fdocs\u002Fintegrations\u002Fagents\">**Agents Support**\u003C\u002Fa>: Seamlessly integrate with popular agent frameworks to build complex AI applications. The gateway seamlessly integrates with [Autogen](https:\u002F\u002Fportkey.wiki\u002Fgh-50), [CrewAI](https:\u002F\u002Fportkey.wiki\u002Fgh-51), [LangChain](https:\u002F\u002Fportkey.wiki\u002Fgh-52), [LlamaIndex](https:\u002F\u002Fportkey.wiki\u002Fgh-53), [Phidata](https:\u002F\u002Fportkey.wiki\u002Fgh-54), [Control Flow](https:\u002F\u002Fportkey.wiki\u002Fgh-55), and even [Custom Agents](https:\u002F\u002Fportkey.wiki\u002Fgh-56).\n- [**Prompt Template Management***](https:\u002F\u002Fportkey.wiki\u002Fgh-57): Create, manage and version your prompt templates collaboratively through a universal prompt playground.\n\u003Cbr\u002F>\u003Cbr\u002F>\n\n\u003Csup>\n* Available in hosted and enterprise versions\n\u003C\u002Fsup>\n\n\u003Cbr>\n\n## Portkey Models\nOpen-source LLM pricing database for 40+ providers - used by the Gateway for cost tracking.\n\n[GitHub](https:\u002F\u002Fgithub.com\u002FPortkey-AI\u002Fmodels) | [Model Explorer](https:\u002F\u002Fportkey.ai\u002Fmodels)\n\n\u003Cbr>\n\n## Cookbooks\n\n### ☄️ Trending\n- Use models from [Nvidia NIM](\u002Fcookbook\u002Fproviders\u002Fnvidia.ipynb) with AI Gateway\n- Monitor [CrewAI Agents](\u002Fcookbook\u002Fmonitoring-agents\u002FCrewAI_with_Telemetry.ipynb) with Portkey!\n- Comparing [Top 10 LMSYS Models](\u002Fcookbook\u002Fuse-cases\u002FLMSYS%20Series\u002Fcomparing-top10-LMSYS-models-with-Portkey.ipynb) with AI Gateway.\n\n### 🚨 Latest\n* [Create Synthetic Datasets using Nemotron](\u002Fcookbook\u002Fuse-cases\u002FNemotron_GPT_Finetuning_Portkey.ipynb)\n* [Use the LLM Gateway with Vercel's AI SDK](\u002Fcookbook\u002Fintegrations\u002Fvercel-ai.md)\n* [Monitor Llama Agents with Portkey's LLM Gateway](\u002Fcookbook\u002Fmonitoring-agents\u002FLlama_Agents_with_Telemetry.ipynb)\n\n[View all cookbooks →](https:\u002F\u002Fportkey.wiki\u002Fgh-58)\n\u003Cbr\u002F>\u003Cbr\u002F>\n\n## Supported Providers\n\nExplore Gateway integrations with [45+ providers](https:\u002F\u002Fportkey.wiki\u002Fgh-59) and [8+ agent frameworks](https:\u002F\u002Fportkey.wiki\u002Fgh-90).\n\n|                                                                                                                            | Provider                                                                                      | Support | Stream |\n| -------------------------------------------------------------------------------------------------------------------------- | --------------------------------------------------------------------------------------------- | ------- | ------ |\n| \u003Cimg src=\"docs\u002Fimages\u002Fopenai.png\" width=35 \u002F>                                                                              | [OpenAI](https:\u002F\u002Fportkey.wiki\u002Fgh-60)                           | ✅       | ✅      |\n| \u003Cimg src=\"docs\u002Fimages\u002Fazure.png\" width=35>                                                                                 | [Azure OpenAI](https:\u002F\u002Fportkey.wiki\u002Fgh-61)               | ✅       | ✅      |\n| \u003Cimg src=\"docs\u002Fimages\u002Fanyscale.png\" width=35>                                                                              | [Anyscale](https:\u002F\u002Fportkey.wiki\u002Fgh-62) | ✅       | ✅      |\n| \u003Cimg src=\"https:\u002F\u002Fupload.wikimedia.org\u002Fwikipedia\u002Fcommons\u002F2\u002F2d\u002FGoogle-favicon-2015.png\" width=35>                           | [Google Gemini](https:\u002F\u002Fportkey.wiki\u002Fgh-63)             | ✅       | ✅      |\n| \u003Cimg src=\"docs\u002Fimages\u002Fanthropic.png\" width=35>                                                                             | [Anthropic](https:\u002F\u002Fportkey.wiki\u002Fgh-64)                     | ✅       | ✅      |\n| \u003Cimg src=\"docs\u002Fimages\u002Fcohere.png\" width=35>                                                                                | [Cohere](https:\u002F\u002Fportkey.wiki\u002Fgh-65)                           | ✅       | ✅      |\n| \u003Cimg src=\"https:\u002F\u002Fassets-global.website-files.com\u002F64f6f2c0e3f4c5a91c1e823a\u002F654693d569494912cfc0c0d4_favicon.svg\" width=35> | [Together AI](https:\u002F\u002Fportkey.wiki\u002Fgh-66)                 | ✅       | ✅      |\n| \u003Cimg src=\"https:\u002F\u002Fwww.perplexity.ai\u002Ffavicon.svg\" width=35>                                                                 | [Perplexity](https:\u002F\u002Fportkey.wiki\u002Fgh-67)                | ✅       | ✅      |\n| \u003Cimg src=\"https:\u002F\u002Fdocs.mistral.ai\u002Fimg\u002Ffavicon.ico\" width=35>                                                               | [Mistral](https:\u002F\u002Fportkey.wiki\u002Fgh-68)                      | ✅       | ✅      |\n| \u003Cimg src=\"https:\u002F\u002Fdocs.nomic.ai\u002Fimg\u002Fnomic-logo.png\" width=35>                                                              | [Nomic](https:\u002F\u002Fportkey.wiki\u002Fgh-69)                             | ✅       | ✅      |\n| \u003Cimg src=\"https:\u002F\u002Ffiles.readme.io\u002Fd38a23e-small-studio-favicon.png\" width=35>                                              | [AI21](https:\u002F\u002Fportkey.wiki\u002Fgh-91)                                    | ✅       | ✅      |\n| \u003Cimg src=\"https:\u002F\u002Fplatform.stability.ai\u002Fsmall-logo-purple.svg\" width=35>                                                   | [Stability AI](https:\u002F\u002Fportkey.wiki\u002Fgh-71)               | ✅       | ✅      |\n| \u003Cimg src=\"https:\u002F\u002Fdeepinfra.com\u002F_next\u002Fstatic\u002Fmedia\u002Flogo.4a03fd3d.svg\" width=35>                                            | [DeepInfra](https:\u002F\u002Fportkey.sh\u002Fgh-92)                               | ✅       | ✅      |\n| \u003Cimg src=\"https:\u002F\u002Follama.com\u002Fpublic\u002Follama.png\" width=35>                                                                  | [Ollama](https:\u002F\u002Fportkey.wiki\u002Fgh-72)                           | ✅       | ✅      |\n| \u003Cimg src=\"https:\u002F\u002Fnovita.ai\u002Ffavicon.ico\" width=35>                                                                         | [Novita AI](https:\u002F\u002Fportkey.wiki\u002Fgh-73)                              | ✅       | ✅      | `\u002Fchat\u002Fcompletions`, `\u002Fcompletions` |\n\n\n> [View the complete list of 200+ supported models here](https:\u002F\u002Fportkey.wiki\u002Fgh-74)\n\u003Cbr>\n\n\u003Cbr>\n\n## Agents\nGateway seamlessly integrates with popular agent frameworks. [Read the documentation here](https:\u002F\u002Fportkey.wiki\u002Fgh-75).\n\n\n| Framework | Call 200+ LLMs | Advanced Routing | Caching | Logging & Tracing* | Observability* | Prompt Management* |\n|------------------------------|--------|-------------|---------|------|---------------|-------------------|\n| [Autogen](https:\u002F\u002Fportkey.wiki\u002Fgh-93)    | ✅     | ✅          | ✅      | ✅   | ✅            | ✅                |\n| [CrewAI](https:\u002F\u002Fportkey.wiki\u002Fgh-94)             | ✅     | ✅          | ✅      | ✅   | ✅            | ✅                |\n| [LangChain](https:\u002F\u002Fportkey.wiki\u002Fgh-95)             | ✅     | ✅          | ✅      | ✅   | ✅            | ✅                |\n| [Phidata](https:\u002F\u002Fportkey.wiki\u002Fgh-96)             | ✅     | ✅          | ✅      | ✅   | ✅            | ✅                |\n| [Llama Index](https:\u002F\u002Fportkey.wiki\u002Fgh-97)             | ✅     | ✅          | ✅      | ✅   | ✅            | ✅                |\n| [Control Flow](https:\u002F\u002Fportkey.wiki\u002Fgh-98) | ✅     | ✅          | ✅      | ✅   | ✅            | ✅                |\n| [Build Your Own Agents](https:\u002F\u002Fportkey.wiki\u002Fgh-99) | ✅     | ✅          | ✅      | ✅   | ✅            | ✅                |\n| \u003Cimg src=\"https:\u002F\u002Fio.net\u002Ffavicon.ico\" width=35> | [IO Intelligence](https:\u002F\u002Fio.net\u002Fintelligence) | ✅ | ✅ |\n\n\u003Cbr>\n\n*Available on the [hosted app](https:\u002F\u002Fportkey.wiki\u002Fgh-76). For detailed documentation [click here](https:\u002F\u002Fportkey.wiki\u002Fgh-100).\n\n\n## Gateway Enterprise Version\nMake your AI app more \u003Cins>reliable\u003C\u002Fins> and \u003Cins>forward compatible\u003C\u002Fins>, while ensuring complete \u003Cins>data security\u003C\u002Fins> and \u003Cins>privacy\u003C\u002Fins>.\n\n✅  Secure Key Management - for role-based access control and tracking \u003Cbr>\n✅  Simple & Semantic Caching - to serve repeat queries faster & save costs \u003Cbr>\n✅  Access Control & Inbound Rules - to control which IPs and Geos can connect to your deployments \u003Cbr>\n✅  PII Redaction - to automatically remove sensitive data from your requests to prevent indavertent exposure \u003Cbr>\n✅  SOC2, ISO, HIPAA, GDPR Compliances - for best security practices \u003Cbr>\n✅  Professional Support - along with feature prioritization \u003Cbr>\n\n[Schedule a call to discuss enterprise deployments](https:\u002F\u002Fportkey.sh\u002Fdemo-13)\n\n\u003Cbr>\n\n\n## Contributing\n\nThe easiest way to contribute is to pick an issue with the `good first issue` tag 💪. Read the contribution guidelines [here](\u002F.github\u002FCONTRIBUTING.md).\n\nBug Report? [File here](https:\u002F\u002Fportkey.wiki\u002Fgh-78) | Feature Request? [File here](https:\u002F\u002Fportkey.wiki\u002Fgh-78)\n\n\n### Getting Started with the Community\nJoin our weekly AI Engineering Hours every Friday (8 AM PT) to:\n- Meet other contributors and community members\n- Learn advanced Gateway features and implementation patterns\n- Share your experiences and get help\n- Stay updated with the latest development priorities\n\n[Join the next session →](https:\u002F\u002Fportkey.wiki\u002Fgh-101) | [Meeting notes](https:\u002F\u002Fportkey.wiki\u002Fgh-102)\n\n\u003Cbr>\n\n## Community\n\nJoin our growing community around the world, for help, ideas, and discussions on AI.\n\n- View our official [Blog](https:\u002F\u002Fportkey.wiki\u002Fgh-78)\n- Chat with us on [Discord](https:\u002F\u002Fportkey.wiki\u002Fcommunity)\n- Follow us on [Twitter](https:\u002F\u002Fportkey.wiki\u002Fgh-79)\n- Connect with us on [LinkedIn](https:\u002F\u002Fportkey.wiki\u002Fgh-80)\n- Read the documentation in [Japanese](.\u002F.github\u002FREADME.jp.md)\n- Visit us on [YouTube](https:\u002F\u002Fportkey.wiki\u002Fgh-103)\n- Join our [Dev community](https:\u002F\u002Fportkey.wiki\u002Fgh-82)\n\u003C!-- - Questions tagged #portkey on [Stack Overflow](https:\u002F\u002Fstackoverflow.com\u002Fquestions\u002Ftagged\u002Fportkey) -->\n\n![Rubeus Social Share (4)](https:\u002F\u002Fgithub.com\u002FPortkey-AI\u002Fgateway\u002Fassets\u002F971978\u002F89d6f0af-a95d-4402-b451-14764c40d03f)\n","Portkey-AI\u002Fgateway 是一个快速且安全的AI网关，能够连接超过1600种语言、视觉、音频和图像模型。该项目采用TypeScript编写，具备极低延迟（\u003C1ms）和轻量级（122kb）的特点，并支持自动重试和故障转移功能，确保服务高可用性。此外，它还提供负载均衡和自定义部署选项，适用于需要大规模处理AI请求的企业级应用场景。无论是快速集成各类语言模型还是构建复杂的人工智能应用，Portkey-AI\u002Fgateway 都能提供稳定高效的支持。",2,"2026-06-11 03:36:52","high_star"]