Platforms & Frameworks

Platforms, frameworks, and tools for building and running AI applications.

hubAPI Gateway

OpenRouter

Unified API gateway for 100+ AI models from OpenAI, Anthropic, Google, Meta, and more. One API key, one format, all models.

apigatewayopenaianthropicgoogleunified
https://openrouter.ai open_in_new
cloudCloud Platform

Amazon Bedrock

AWS managed service providing access to foundation models from Amazon, Anthropic, Meta, and others. Enterprise-grade with built-in security and compliance.

awsamazoncloudenterprisebedrockcompliance
https://aws.amazon.com/bedrock open_in_new
cloudCloud Platform

Microsoft Azure AI

Cloud platform offering OpenAI models, open-source models, and AI services. Enterprise compliance with SOC 2, HIPAA, and GDPR support.

azuremicrosoftcloudenterpriseopenaicompliance
https://azure.microsoft.com/en-us/products/ai-services open_in_new
groupsPlatform

Hugging Face

The largest open-source AI community and platform. Host, share, and deploy models, datasets, and Spaces. The GitHub of machine learning.

open sourcecommunitymodelsdatasetsspaceshub
https://huggingface.co open_in_new
play_circlePlatform

Replicate

Cloud platform for running and fine-tuning open-source AI models. Simple API — deploy models without managing infrastructure.

cloudinferencefine-tuningapideploy
https://replicate.com open_in_new
codeIDE

Cursor

AI-first code editor built on VS Code. Integrates with Claude, GPT-4, and other models for code generation, editing, and chat.

idevscodecode generationclaudegpteditor
https://cursor.com open_in_new
desktop_windowsLocal Inference

LM Studio

Desktop application for running open-source LLMs locally. User-friendly GUI for downloading, managing, and chatting with models offline.

localdesktopguiofflineopen sourceinference
https://lmstudio.ai open_in_new
terminalLocal Inference

llama.cpp

High-performance C/C++ inference engine for LLMs. Runs models on CPU and GPU with quantization. Powers many local AI tools including Ollama.

c++inferencecpugpuquantizationlocalollama
https://github.com/ggerganov/llama.cpp open_in_new
memoryLocal Inference

MLX

Apple Silicon ML framework by Apple. Run and fine-tune models natively on Mac with unified memory. Optimized for M-series chips.

applemacsiliconlocalfine-tuningmlx
https://github.com/ml-explore/mlx open_in_new
tuneFine-tuning

Unsloth

Platform for fine-tuning LLMs 2-5x faster with 80% less memory. Supports LoRA and QLoRA for efficient model customization.

fine-tuningloraqloratrainingoptimization
https://unsloth.ai open_in_new
monitoringObservability

Langfuse

Open-source observability platform for LLM applications. Trace, evaluate, and debug your AI pipelines with detailed analytics.

observabilitytracinganalyticsdebuggingopen source
https://langfuse.com open_in_new
scienceEvaluation

Promptfoo

Open-source tool for testing and evaluating LLM prompts. Compare models, run red-team tests, and catch regressions in prompt quality.

testingevaluationred-teampromptsqualityopen source
https://promptfoo.dev open_in_new
auto_fix_highFramework

DSPy

Framework for programming — not prompting — foundation models. Define signatures and let the compiler optimize your prompts automatically.

frameworkcompileroptimizationsignaturesprogramming
https://github.com/stanfordnlp/dspy open_in_new
spaFramework

Spring AI

Spring framework integration for building AI-powered applications in Java. Supports chat models, embeddings, vector stores, and function calling.

springjavaframeworkenterpriseembeddingsfunction calling
https://spring.io/projects/spring-ai open_in_new
storefrontMarketplace

SkillsMP

Open marketplace for AI agent skills — reusable SKILL.md instructions that teach coding assistants (Claude Code, Codex, ChatGPT) how to handle specific tasks. Browse 1.8M+ community-built skills by keyword, occupation, or creator.

agent skillsSKILL.mdmarketplaceClaude CodeCodexopen sourcecommunityautomation
https://skillsmp.com open_in_new