AI Infrastructure

AI legal assistant for law firms specializing in research, drafting, and contract review

Alibaba's Qwen 3 — open-source frontier model family with hybrid thinking mode and strong multilingual performance.

Desktop app to run LLMs locally on your Mac or PC — download and chat with Llama, Mistral, Phi and hundreds of models offline.

Open-source offline AI assistant — run ChatGPT-like conversations entirely on your device with full privacy.

Supabase's AI features — vector embeddings, pgvector search and AI SQL assistant built into the open-source Firebase alternative.

AI medical documentation tool that converts patient-physician conversations into structured clinical notes.

AI model gateway and workspace — compare, test and use 200+ AI models in one unified interface with team collaboration.

Open-source visual LLM flow builder — drag and drop to create RAG chatbots, agents and AI workflows without code.

W&B's LLM evaluation and observability tool — trace, evaluate and improve AI application quality in production.

The data framework for LLM applications — connect LLMs to your data with RAG, agents and advanced query pipelines.

AI legal platform for contract review, due diligence, and negotiation with deep legal reasoning.

Open-source LLM observability — trace, evaluate and debug your AI applications with detailed prompt analytics.

Meta's Llama 4 — open-weight multimodal frontier model with MoE architecture, available for free commercial use.

Free URL-to-LLM-markdown converter — prefix any URL with r.jina.ai and get clean text content perfect for AI processing.

One-click installer for any AI app — run Stable Diffusion, Llama, ComfyUI and thousands of AI tools locally without tech knowledge.

Open-source LLM app development platform — build AI agents, RAG pipelines and chatbots with a visual workflow builder.

Mistral AI's API — fast, efficient European LLMs including Mistral Large, Small and open-weight models via simple API.

AI tool that instantly differentiates any reading material to match students' exact reading levels.

Groq's LPU inference — the fastest LLM inference in the world, running Llama and Mistral at 500+ tokens/second.

Microsoft's Phi-4 — small language model punching above its weight with outstanding reasoning despite running on-device.

AI model routing API — one API key for 300+ LLMs including GPT, Claude, Gemini, Llama and Mistral with automatic fallbacks.

Fast inference for open-source models — run Llama, Mistral, FLUX and 200+ models at the fastest speeds via API.

Run AI models in the cloud with one line of code — 50,000+ models including FLUX, Stable Diffusion and Whisper via API.

AI gateway for production LLM apps — load balancing, fallbacks, caching, observability and guardrails in one platform.

Open-source vector database — store embeddings, run semantic search and build RAG applications at scale.

Cerebras wafer-scale chip inference — run Llama models at 2,000+ tokens/second, the world's fastest AI inference.

Cohere's enterprise AI platform — RAG, embeddings, Command models and on-premise deployment for business applications.

Anthropic's API for Claude models — build AI applications with Claude 3.5 Sonnet, Haiku and Opus via a simple REST API.

High-performance open-source vector database written in Rust — fast, accurate and production-ready for RAG apps.