AI platforms, gateways, and APIs
This category is the glue: model APIs, app builders, research notebooks, gateways. OpenAI is the API most tutorials still assume. Dify is an open-source LLM app builder you can cloud or self-host. NotebookLM was renamed Gemini Notebook in July 2026. If you need a coding IDE, go to Dev Tools instead of shopping here.
View rankingsBrowse alternativesBrowse by tags
Dify vs building on the OpenAI API?
Use Dify when you want workflows, RAG, and an admin UI without writing the whole stack. Use the raw API when you already have an app and only need inference.
Is NotebookLM still a product?
Yes. Google renamed it Gemini Notebook on 2026-07-16. Same research tool, more Gemini-app integration. Search both names.
Cloud or self-host?
Cloud is faster to try. Self-host if data cannot leave your network. Dify's community edition is the usual self-host path in this hub.
All AI platforms, gateways, and APIs tools (35)
Open any tool below for details, features, and pricing.
- 9RouterFree MIT AI router and token saver that connects Claude Code, Codex, Cursor, and Cline to 40+ providers with auto-fallback and RTK token compression.
- AnySearchPrivacy-first AI search infrastructure for agents: one API and MCP server unifying web and vertical-domain search with structured results that cut token waste.
- ApodexSelf-evolving heavy-duty solver for deep research and agentic work. Splits reasoner from verifier, runs a trained Agent Team, and ships open weights plus the FrontierAgent CLI research harness.
- CheaperInferenceCheaperInference resells discounted AI inference through one OpenAI-compatible API, cutting model costs by up to 30% with usage-based billing, no contract.
- Claude AcademyAnthropic's official learning platform. 300+ free courses and tutorials across Claude.ai, Cowork, Claude Code, Tag, and the Platform API, plus an AI Fluency.
- Claude SecurityAnthropic's enterprise code security scanning, now powered by Claude Mythos 5 in public beta, tracing data across files to report CWE categories, severity, and suggested fixes.
- Cloudflare AI GatewayLLM proxy on Cloudflare. Paid logs 10M/gateway; extra requests $0.05/million. Not an unmetered endpoint.
- CozeByteDance agent studio. Coze Studio on GitHub is Apache-2.0 with 21,459 stars. Not a free-for-anyone slogan.
- DifyOpen-source agent workflows. Cloud: Sandbox $0, Professional $59, Team $159 per workspace/month. Community self-host is $0.
- ExLlamaV3An efficient local LLM inference library built around the EXL3 quantization format, running large open models on consumer GPUs with tensor and expert parallelism, and powering the TabbyAPI server.
- FirecrawlWeb-to-LLM API. AGPL-3.0 self-host. Not the old $16/$83/$333-only table.
- Free Claude CodeMIT local gateway that points Claude Code, Codex, Pi, OpenCode, or Cline at 48 providers.
- GarakNVIDIA's open-source LLM vulnerability scanner with dozens of plugins and thousands of prompts for probing jailbreaks, prompt injection, and data leakage.
- GMI CloudAI-native GPU cloud. No invented seat.
- InsForgeOpen-source agent-native backend: Postgres, auth, storage, edge functions, compute, hosting, and a model gateway driven by CLI and MCP.
- LangChainMIT agent framework. Hosted traces are LangSmith Developer $0 / Plus $39 per seat. Not LangServe-as-the-product.
- LlamaIndexDocument agents plus LlamaParse. 1,000 credits = $1.25. OSS framework is MIT.
- MCP FrameworkTypeScript kit for MCP servers. Software $0. Not a 5-minute Claude Desktop-only tutorial.
- MCP India StackMCP India Stack is a MIT offline-first MCP server with 76 Indian tax, identity, and legal tools for agents.
- MiniMaxShanghai lab behind Hailuo, M2.5, and the 2026 M3/H3 coding models. TalkAPI M3 is $0.24/$0.98 per 1M tokens.
- mlx-servemlx-serve is a MIT Zig inference server for Apple Silicon that runs MLX and GGUF locally with OpenAI and Anthropic APIs, no Python.
- NVIDIA NeMo GuardrailsNVIDIA's open-source toolkit for adding programmable guardrails to LLM conversational applications, with Colang flows for input, dialog, retrieval, and output rails.
- NInferNInfer is an Apache-2.0 C++/CUDA engine for selected Qwen checkpoints on a single RTX 5090, with OpenAI- and Anthropic-compatible local APIs.
- NotebookLMGoogle's source-grounded research notebook, now presented as Gemini Notebook at notebook.google. Analyzes your files and turns them into audio, video, and more.
- OllamaLocal models plus Ollama cloud. Not a 30-minute Llama3 tutorial.
- oMLXoMLX is an Apache-2.0 Apple Silicon inference server with SSD KV cache and a macOS menu bar app for coding agents.
- OmniRouteFree MIT AI gateway with one endpoint, 339 providers, 90+ free tiers, quota-aware fallback, and token compression for Claude Code, Codex, and Cursor.
- OpenAISan Francisco lab behind ChatGPT and the GPT API. Current flagship is GPT-5.6 Sol; GPT-5.2 remains the previous frontier snapshot.
- OpenGradientDecentralized L1 for verifiable AI: ZKML, TEE, Vanilla, HACA, PIPE, x402, Model Hub. No invented dollar table.
- OpenRouterUnified LLM router. Not a GPT-4o catalog page.
- RAGFlowOpen-source RAG plus agents. Cloud signup exists; no public dollar table. Not a generic document scanner.
- SiliconFlowChina inference API for open models. Official site siliconflow.cn. Qwen3-8B listed at ¥0.27/¥0.88 per 1M tokens. Skip unverified 6M-user claims.
- skills.qiaomu.aiChinese-language Claude Code Skill recommendation site by Yang Xiang Qiaomu (向阳乔木): a curated catalog of popular Skills with one-click install guidance for d.
- SubmitlistFree product-launch tracker that manages your submissions to 300+ directories and launch platforms, with an MCP server so AI agents handle the paperwork.
- ToAPIsOpenAI-compatible AI API gateway: one base URL and key for 60+ models from GPT, Claude, Gemini, DeepSeek, Sora, and VEO, with routing, failover, and unified.