A self-updating local repository containing 29,000+ AI skills, 23 MCP servers, 3,348 IDE rules, system prompts, API references, and no-code platform documentation—designed for AI developers and agents to work offline without subscriptions.
A massive, self-updating local archive of AI tools — 11,000+ agent skills, 240+ MCP servers, 2,200+ IDE rules (Cursor/Cline), and 30+ system prompt collections. One repo to rule them all.
README
AI Workspace Archive
The most comprehensive self-hostable AI developer toolbox on GitHub.
29,746 agent skills & files | 23 MCP servers | 3,348 IDE rules | 8 system prompt collections | 12 API providers | 15 no-code platforms | 1,400+ public APIs -- 174 curated upstream repos, all free
One repository. Everything an AI developer needs. Offline. No subscriptions. No gatekeeping.
Table of Contents
| # | Pillar | Scale | Path |
|---|---|---|---|
| 1 | AI Skills and Prompt Library | 29,746 files from curated repos | /ai_skills_library/ |
| 2 | MCP Server Repository | 23 repos across 11 categories | /mcps/ |
| 3 | IDE Context Rules | 3,348 AI rules for every stack | /ide_rules/ |
| 4 | System Prompts and Frameworks | 8 repos across 4 categories | /system_prompts/ |
| 5 | AI API Providers Reference | 12 providers, 50+ models | /api_providers/ |
| 6 | No-Code and Visual Workflow Builders | 15 repos, 29 cloned directories | /nocode_platforms/ |
| 7 | Public APIs Directory | 1,400+ APIs across 40+ categories | /public_apis/ |
| 8 | AI Agents & Advanced Tooling | Autonomous agents, frameworks & scraping | /ai_agents/ |
| - | Quick Start | Copy-paste commands for each pillar | - |
| - | Auto-Sync System | VPS bot + GitHub Actions | - |
| - | Archive Stats | Full breakdown | - |
Pillar 1 -- AI Skills and Prompt Library
/ai_skills_library/-- 29,746 structured YAML-frontmatted skill files, deep-categorized across 20 sub-domains, sourced from curated GitHub repositories
Every file follows the same structure: a YAML frontmatter block declaring the skill name, recommended model, input format, and usage context -- followed by the actual prompt content. Drop any file directly into Claude, Cursor, Gemini, Copilot, or any AI assistant as a system prompt.
ai_skills_library/skills/
|
+-- development/
| +-- frontend/ React, Vue, Angular, Next.js, Svelte, CSS, HTML, animations
| +-- backend/ FastAPI, Express, Django, Rails, Go APIs, gRPC, REST design
| +-- database/ SQL optimization, schema design, migrations, indexing
| +-- devops/ Docker, Kubernetes, Terraform, CI/CD, GitHub Actions, AWS
| +-- mobile/ React Native, Flutter, Swift, Kotlin, iOS, Android
| +-- security/ Pen testing, code review, OWASP, vulnerability analysis
| +-- ai-ml/ LLM integration, embeddings, RAG, fine-tuning, agents
|
+-- productivity/
| +-- writing/ Emails, reports, documentation, blog posts, READMEs
| +-- analysis/ Research synthesis, data analysis, competitor research
| +-- planning/ Project specs, task breakdown, sprint planning, roadmaps
| +-- communication/ Meeting summaries, presentations, Slack messages
| +-- finance/ Financial modeling, budget analysis, investment memos
|
+-- creative/
| +-- design/ UI/UX briefs, design systems, Figma instructions
| +-- marketing/ Ad copy, landing pages, social media, brand voice
| +-- multimedia/ Video scripts, podcast outlines, image prompts
|
+-- other/
+-- uncategorized/ Domain-specific skills: legal, medical, education, science
| Folder | Original Repository | Focus Area |
|---|---|---|
agentscope-skills |
modelscope/agentscope | Multi-agent framework skills |
alirezarezvani-skills |
Community contribution | Persian-language AI skills |
anthropics-skills |
anthropics/anthropic-cookbook | Official Anthropic prompt recipes |
apify-skills |
apify/actors-mcp-server | Web scraping and automation |
awesome-copilot |
github/awesome-copilot | GitHub Copilot instructions |
factory-skills |
Community contribution | Software factory workflows |
flare-skills |
Community contribution | Flare platform skills |
gmh5225-awesome-skills |
Community curation | Mixed domain skills |
heilcheng-agent-skills |
Community contribution | Chinese-language agent skills |
hf-skills |
huggingface/agents-course | HuggingFace AI agent course |
integralist-skills |
Community contribution | DevOps and SRE focused skills |
karanb-claude-skills |
Community contribution | Claude-specific skill patterns |
obsidian-skills |
Community contribution | Obsidian PKM productivity skills |
openai-skills |
openai/openai-cookbook | Official OpenAI examples |
openclaw-skills |
Community contribution | Legal and contract AI skills |
planning-with-files |
Community contribution | File-based planning skills |
promptdeploy |
Community contribution | Prompt deployment workflows |
shajith-skills |
Community contribution | Full-stack dev skills |
sickn33-awesome-skills |
Community curation | The original comprehensive skill list |
sugarforever-skills |
Community contribution | Data science and ML skills |
voltagent-agent-skills |
voltagent/voltagent | VoltAgent framework skills |
wordpress-skills |
Community contribution | WordPress and CMS skills |
yoriiis-skills |
Community contribution | JavaScript ecosystem skills |
Frontend (development/frontend/)
- React component generation with proper hooks, context, and TypeScript
- Next.js App Router pages with RSC, server actions, and metadata
- Vue 3 Composition API components with Pinia state management
- CSS animation and glassmorphism design patterns
- Accessibility audit and ARIA implementation
- Performance optimization: lazy loading, code splitting, bundle analysis
Backend (development/backend/)
- FastAPI endpoint generation with Pydantic models and async patterns
- Express.js REST API scaffolding with middleware and error handling
- Django views, serializers, and ORM query optimization
- GraphQL schema design and resolver implementation
- gRPC service definition and implementation
- API design review and OpenAPI documentation
Database (development/database/)
- PostgreSQL query optimization and index strategy
- MongoDB schema design and aggregation pipelines
- Database migration scripting and rollback plans
- Redis caching layer design
- SQL to NoSQL migration planning
DevOps (development/devops/)
- Dockerfile multi-stage build optimization
- Kubernetes manifest generation (Deployments, Services, Ingress, HPA)
- Terraform module writing for AWS/GCP/Azure
- GitHub Actions workflow construction
- Helm chart creation and values configuration
- Infrastructure cost analysis
Security (development/security/)
- OWASP Top 10 vulnerability checklist scanning
- Dependency audit and CVE identification
- JWT authentication flow design
- Penetration testing plan generation
- Secure code review with specific code pattern checks
AI/ML (development/ai-ml/)
- RAG pipeline design and implementation
- LLM evaluation framework setup
- Embedding generation and vector store integration
- Fine-tuning dataset preparation
- Agent reasoning chain construction
- Multi-modal prompt engineering
Writing (productivity/writing/)
- Technical documentation generation from code
- README.md creation with badges, TOC, examples
- Executive summary distillation from long documents
- Blog post drafting with SEO structure
- Email drafting for various tones (formal, startup, sales)
- API reference documentation
Analysis (productivity/analysis/)
- Research paper summarization with key findings extraction
- Competitor analysis with SWOT framework
- Market sizing and TAM/SAM/SOM breakdown
- Data interpretation and insight generation
- Literature review synthesis
- Survey result analysis
Planning (productivity/planning/)
- Project spec document generation
- Sprint planning and story point estimation
- OKR and KPI framework setup
- Risk assessment matrix construction
- Roadmap timeline creation
- Resource allocation planning
Finance (productivity/finance/)
- Financial model construction in tabular format
- Investor memo drafting
- Budget variance analysis
- Unit economics calculation (CAC, LTV, MRR)
- Cash flow projection
Design (creative/design/)
- Design system token specification (colors, typography, spacing)
- Component UI brief writing for developers
- UX research interview script generation
- Figma component naming convention setup
- Brand identity guidelines documentation
Marketing (creative/marketing/)
- Landing page copy with CTA optimization
- Google/Meta ad copy variants (A/B sets)
- Social media calendar and content planning
- Email campaign sequence writing
- Product launch announcement drafting
- Brand voice guide creation
Multimedia (creative/multimedia/)
- YouTube video script with hook, body, CTA
- Podcast episode outline and question generation
- Midjourney/DALL-E prompt engineering
- Short-form video caption writing
- Thumbnail concept descriptions
Pillar 2 -- MCP Server Repository
/mcps/-- 23 fully cloned Model Context Protocol server repositories across 11 categories -- all installable, all production-ready, connecting directly to Claude Desktop, Cursor, Windsurf, or any MCP-compatible client
| Repo Folder | GitHub | Description |
|---|---|---|
servers |
modelcontextprotocol/servers | Official MCP reference server implementations |
python-sdk |
modelcontextprotocol/python-sdk | Python SDK for building MCP servers and clients |
typescript-sdk |
modelcontextprotocol/typescript-sdk | TypeScript/Node.js SDK |
java-sdk |
modelcontextprotocol/java-sdk | Java SDK for JVM-based MCP servers |
kotlin-sdk |
modelcontextprotocol/kotlin-sdk | Kotlin SDK for Android/JVM |
csharp-sdk |
modelcontextprotocol/csharp-sdk | C# / .NET SDK |
rust-sdk |
modelcontextprotocol/rust-sdk | Rust SDK for high-performance servers |
inspector |
modelcontextprotocol/inspector | Visual debugging tool for MCP servers |
specification |
modelcontextprotocol/specification | Official MCP protocol specification |
docs |
modelcontextprotocol/modelcontextprotocol.io | Official documentation site |
create-typescript-server |
modelcontextprotocol/create-typescript-server | Scaffolder for new TS MCP servers |
| Repo Folder | GitHub | Stars | Description |
|---|---|---|---|
playwright-mcp |
microsoft/playwright-mcp | 12K+ | Microsoft official Playwright MCP -- control any browser |
mcp-playwright |
executeautomation/mcp-playwright | 800+ | Playwright test automation via MCP |
mcp-server-browserbase |
browserbase/mcp-server-browserbase | 500+ | Cloud browser automation with Browserbase |
chrome-devtools-mcp |
AgentDeskAI/browser-tools-mcp | 500+ | Access Chrome DevTools Protocol via MCP |
| Repo Folder | GitHub | Stars | DB / Service |
|---|---|---|---|
supabase-mcp |
supabase-community/supabase-mcp | 1K+ | Supabase (PostgreSQL + Auth + Storage) |
mcp-snowflake-server |
isaacwasserman/mcp-snowflake-server | 300+ | Snowflake data warehouse |
mcp-bigquery-server |
ergut/mcp-bigquery-server | 200+ | Google BigQuery |
mcp-mongo-server |
mongodb-developer/mcp-mongo-server | 400+ | MongoDB document database |
mcp-redis |
redis/mcp-redis | 300+ | Redis key-value cache |
mcp-neo4j |
neo4j-contrib/mcp-neo4j | 200+ | Neo4j graph database |
mcp-clickhouse |
ClickHouse/mcp-clickhouse | 150+ | ClickHouse analytics DB |
mcp-mysql-server |
Community | 100+ | MySQL database |
dbhub |
bytebase/dbhub | 400+ | Universal DB gateway (multi-DB support) |
mcp-database-server |
Community | - | Universal database MCP |
mysql_mcp_server |
Community | - | Secondary MySQL implementation |
mcp-sqlite-server |
Community | - | Local SQLite DB via MCP |
| Repo Folder | GitHub | Stars | Platform |
|---|---|---|---|
aws-mcp-server |
alexei-led/aws-mcp-server | 500+ | AWS -- EC2, S3, Lambda, IAM, CloudWatch |
kubernetes-mcp-server |
manusa/kubernetes-mcp-server | 300+ | Kubernetes cluster management (Java) |
mcp-k8s-go |
strowk/mcp-k8s-go | 400+ | Kubernetes management (Go, fast) |
| Repo Folder | GitHub | Stars | Service |
|---|---|---|---|
exa-mcp-server |
exa-labs/exa-mcp-server | 600+ | Exa -- AI-native semantic search |
tavily-mcp |
tavily-ai/tavily-mcp | 300+ | Tavily -- research-focused search API |
brave-search-mcp |
Community | 200+ | Brave Search -- private web search |
serpapi-mcp-server |
Community | 100+ | SerpAPI -- Google/Bing/YouTube results |
actors-mcp-server |
apify/actors-mcp-server | 400+ | Apify -- web scraping actors as tools |
| Repo Folder | GitHub | Stars | What It Does |
|---|---|---|---|
mcp-filesystem-server |
modelcontextprotocol/servers | 15K+ | Read/write local files and directories |
mcp-server-commands |
g0t4/mcp-server-commands | 200+ | Run shell commands via MCP |
mcp-youtube |
suekou/mcp-youtube | 150+ | YouTube transcript and video data |
filesystem_servers |
Community | - | Additional filesystem integration wrappers |
delimit-mcp-server |
delimit-ai/delimit-mcp-server | 10+ | API governance, persistent memory, multi-model context, policy enforcement |
| Repo Folder | GitHub | Stars | Language / Purpose |
|---|---|---|---|
python-sdk |
modelcontextprotocol/python-sdk | 5K+ | Build MCP servers in Python |
typescript-sdk |
modelcontextprotocol/typescript-sdk | 5K+ | Build MCP servers in TypeScript |
java-sdk |
modelcontextprotocol/java-sdk | 500+ | Build MCP servers in Java |
kotlin-sdk |
modelcontextprotocol/kotlin-sdk | 300+ | Build MCP servers in Kotlin |
csharp-sdk |
modelcontextprotocol/csharp-sdk | 400+ | Build MCP servers in C# |
rust-sdk |
modelcontextprotocol/rust-sdk | 300+ | Build MCP servers in Rust |
mcp-agent |
lastmile-ai/mcp-agent | 500+ | Agent orchestration framework using MCP |
mcp-framework |
QuantGeekDev/mcp-framework | 300+ | Framework for building MCP servers quickly |
create-typescript-server |
modelcontextprotocol/create-typescript-server | 400+ | CLI scaffolder for new TypeScript MCP servers |
mcphost |
Community | - | MCP Host implementation |
inspector |
modelcontextprotocol/inspector | - | Inspector tools |
openai-agents-python |
Community | - | Python agents integration |
| Repo Folder | GitHub | Stars | Description |
|---|---|---|---|
github-mcp-server |
github/github-mcp-server | 5K+ | Official GitHub MCP -- PRs, issues, repos |
git-mcp |
idosal/git-mcp | 400+ | Git operations: commit, branch, diff, blame |
mcp-filesystem-server |
modelcontextprotocol/servers | 15K+ | File-level operations tied to git workflows |
mcp-package-docs |
upstash/mcp-package-docs | 200+ | Fetch npm/PyPI package documentation |
mcp-server-commands |
g0t4/mcp-server-commands | 200+ | Run git commands as MCP tools |
| Repo Folder | GitHub | Stars | Description |
|---|---|---|---|
awesome-mcp-list |
wong2/awesome-mcp-servers | 5K+ | Curated list of 500+ MCP servers |
mcpm |
mcpm/mcpm | 300+ | MCP package manager -- install servers like npm |
curated_awesome_lists |
Community | - | Additional curated references |
| Repo Folder | Category | Description |
|---|---|---|
mcp-agent |
Discovery | Agent orchestration via MCP |
mcp-hfspace |
HuggingFace | Access HuggingFace Spaces from MCP |
mcp-k8s-go |
Infra | Kubernetes management via AI |
mcp-snowflake-server |
Data | Snowflake database intel |
mcphost |
Server | Hosting bindings |
openai-agents-python |
Python | SDK integration layer |
Includes 27 deeply specialized MCP implementations ranging across APIs, niche data platforms, communication bridges (Slack/Teams), and local productivity bridges:
- Miscellaneous standalone servers (25 repos)
- Communication & Collaboration servers (1 repo)
- Additional Productivity tools (1 repo)
These can be browsed safely via the mcps/misc_specialized/ folder.
1. Find the server you want:
ls mcps/databases/ # List all database servers
ls mcps/browser_automation/ # List all browser servers
2. Install it:
cd mcps/databases/supabase-mcp
npm install # or: pip install -e . / cargo build
3. Add to Claude Desktop config (~/Library/Application Support/Claude/claude_desktop_config.json):
{
"mcpServers": {
"supabase": {
"command": "node",
"args": ["/path/to/mcps/databases/supabase-mcp/build/index.js"],
"env": { "SUPABASE_URL": "...", "SUPABASE_KEY": "..." }
}
}
}
4. Restart Claude Desktop -- the tool appears automatically.
Pillar 3 -- IDE Context Rules
/ide_rules/-- 3,348 AI editor instruction files for Cursor, Windsurf, Cline, and GitHub Copilot -- covering every major language, framework, and tech stack
Drop one file into your project root and your AI editor instantly understands your architecture, coding conventions, testing standards, and style preferences.
| File Type | Compatible With | Where to Place |
|---|---|---|
.cursorrules |
Cursor IDE | Project root |
.mdc files |
Cursor (newer format) | .cursor/rules/*.mdc |
.clinerules |
Cline VS Code Extension | Project root |
.github/copilot-instructions.md |
GitHub Copilot | .github/ folder |
SYSTEM_PROMPT.md |
Any AI chat | Use as system prompt |
| Rule File | Stack | What It Enforces |
|---|---|---|
react-typescript.cursorrules |
React + TypeScript | hooks-only patterns, strict types, no class components |
nextjs-app-router.cursorrules |
Next.js 14+ | App Router, RSC, server actions, metadata API |
nextjs-tailwind.mdc |
Next.js + Tailwind | utility-first, no custom CSS, mobile-first |
vue3-typescript.cursorrules |
Vue 3 + TS | Composition API, script setup, Pinia, Vite |
nuxt3.cursorrules |
Nuxt 3 | auto-imports, Nitro, composables, useAsyncData |
svelte5.cursorrules |
Svelte 5 | runes syntax, $state, $derived, $effect |
angular17.cursorrules |
Angular 17+ | standalone components, signals, new control flow |
remix.cursorrules |
Remix | loaders, actions, progressive enhancement |
astro.cursorrules |
Astro | island architecture, content collections, SSG |
solid-js.cursorrules |
SolidJS | fine-grained reactivity, no virtual DOM patterns |
| Rule File | Stack | What It Enforces |
|---|---|---|
python-fastapi.cursorrules |
FastAPI + Python | Pydantic v2, async/await, dependency injection |
python-django.cursorrules |
Django | ORM patterns, CBVs vs FBVs, signals |
python-flask.cursorrules |
Flask | blueprints, application factory pattern |
nodejs-express.cursorrules |
Node.js + Express | middleware chain, error handling, REST conventions |
go-gin.cursorrules |
Go + Gin | idiomatic Go, error propagation, zero-alloc patterns |
rust-axum.cursorrules |
Rust + Axum | ownership patterns, tower middleware, async traits |
java-spring.cursorrules |
Spring Boot | DI, JPA, REST with proper HTTP semantics |
ruby-rails.cursorrules |
Ruby on Rails | convention over config, ActiveRecord, REST |
php-laravel.cursorrules |
Laravel | Eloquent, Blade, service providers |
elixir-phoenix.cursorrules |
Elixir + Phoenix | functional patterns, LiveView, contexts |
| Rule File | Stack | What It Enforces |
|---|---|---|
react-native.cursorrules |
React Native | platform-specific code, performance, navigation |
expo.cursorrules |
Expo | managed workflow, EAS, expo-router |
flutter.cursorrules |
Flutter | widget tree, state management, null safety |
swift-swiftui.cursorrules |
Swift + SwiftUI | MVVM, Combine, async/await, ViewModels |
kotlin-android.cursorrules |
Kotlin Android | Coroutines, Jetpack, MVVM, Room |
| Rule File | Stack | What It Enforces |
|---|---|---|
postgresql.cursorrules |
PostgreSQL | indexing strategy, EXPLAIN plans, CTEs |
prisma.cursorrules |
Prisma ORM | schema design, migrations, query optimization |
drizzle.cursorrules |
Drizzle ORM | type-safe queries, schema-first, edge-compatible |
mongodb.cursorrules |
MongoDB | schema design, aggregation, indexing |
supabase.cursorrules |
Supabase | RLS policies, Edge Functions, realtime |
firebase.cursorrules |
Firebase | Firestore rules, Cloud Functions, Auth |
| Rule File | Stack | What It Enforces |
|---|---|---|
langchain.cursorrules |
LangChain | LCEL, chains, agents, memory patterns |
langgraph.cursorrules |
LangGraph | stateful agents, graph nodes, human-in-loop |
llamaindex.cursorrules |
LlamaIndex | RAG pipelines, indices, query engines |
openai-api.cursorrules |
OpenAI SDK | streaming, function calling, token management |
anthropic-api.cursorrules |
Anthropic SDK | tool use, streaming, prompt caching |
huggingface.cursorrules |
HuggingFace | transformers, datasets, PEFT, inference |
pytorch.cursorrules |
PyTorch | tensor ops, training loops, CUDA, DataLoaders |
| Rule File | Stack | What It Enforces |
|---|---|---|
dockerfile.cursorrules |
Docker | multi-stage builds, layer caching, non-root users |
kubernetes.cursorrules |
Kubernetes | resource limits, probes, RBAC, network policies |
terraform.cursorrules |
Terraform | modules, state management, naming conventions |
github-actions.cursorrules |
GitHub Actions | workflow reuse, secrets, matrix builds, caching |
helm.cursorrules |
Helm | chart structure, values hierarchy, templating |
# 1. Find your stack
ls ide_rules/cursor/ | grep -i next
# 2. Copy to your project
cp ide_rules/cursor/nextjs-app-router.cursorrules /your-project/.cursorrules
# 3. Open Cursor -- AI is instantly context-aware
# For newer Cursor MDC format:
mkdir -p /your-project/.cursor/rules
cp ide_rules/cursor/nextjs-app-router.mdc /your-project/.cursor/rules/
# For GitHub Copilot:
cp ide_rules/copilot/nextjs.md /your-project/.github/copilot-instructions.md
Pillar 4 -- System Prompts and Frameworks
/system_prompts/-- 8 carefully curated repositories spanning leaked production system prompts, prompt engineering masterclasses, prompt management tools, testing frameworks, and AI safety guardrails
[!WARNING] Legal & Educational Disclaimer: The leaked proprietary system prompts below (e.g., Claude Code, Cursor, Devin) are archived strictly for educational research, syntax analysis, and safe-guard testing. We do not claim ownership of these prompt strings. Users are responsible for adhering to all creator ToS and DMCA regulations.
| Folder | Repository | Stars | What is Inside |
|---|---|---|---|
system-prompts-and-models-of-ai-tools |
x1xhlol/system-prompts-and-models-of-ai-tools | 111K+ | Claude Code, Cursor, Windsurf, Devin, v0, Perplexity, Notion AI, GitHub Copilot, Lovable, Bolt |
thebigpromptlibrary |
0xeb/TheBigPromptLibrary | 5K+ | Massive curated library of GPT and Claude system prompts |
chatgpt_system_prompt |
LouisShark/chatgpt_system_prompt | 8K+ | Hundreds of leaked ChatGPT custom GPT instructions |
What you get from the 111K-star repo: Full system prompts for Claude Code, Cursor, Windsurf/Cascade, Devin, v0 (Vercel), Perplexity, Notion AI, GitHub Copilot, Lovable, Bolt.new, and Manus.
| Folder | Repository | Stars | Content |
|---|---|---|---|
awesome-chatgpt-prompts |
f/awesome-chatgpt-prompts | 115K+ | 157 expert role prompts |
awesome-aipromptlib |
Community | 5K+ | Curated prompt patterns |
Prompt_Engineering |
NirDiamant/Prompt_Engineering | 8K+ | 22 hands-on Jupyter tutorials: zero-shot, few-shot, CoT, ReAct, ToT |
| Folder | Repository | Stars | What It Does |
|---|---|---|---|
promptfoo |
promptfoo/promptfoo | 5K+ | Run automated test suites against your prompts. Compare model outputs, catch regressions |
helicone |
helicone/helicone | 2K+ | LLM observability: request logging, cost tracking, latency, prompt versioning |
langfuse |
langfuse/langfuse | 7K+ | LLM engineering platform: traces, evaluations, prompt management |
| Folder | Repository | Stars | What It Does |
|---|---|---|---|
NeMo-Guardrails |
NVIDIA/NeMo-Guardrails | 4K+ | NVIDIA toolkit for programmable guardrails via Colang DSL |
guardrails |
guardrails-ai/guardrails | 4K+ | Validate and fix LLM outputs against schemas |
rebuff |
protectai/rebuff | 1K+ | Self-hardening prompt injection detector |
Pillar 5 -- AI API Providers Reference
/api_providers/-- Complete offline reference for 12 AI API providers: every model listed, full pricing tables, SDK install commands, and code examples
| Category | Winner | Details | Runner Up |
|---|---|---|---|
| Cheapest Input | DeepSeek V3 | $0.28/M tokens | Gemini 2.5 Flash-Lite ($0.10/M) |
| Best Free Tier | DeepSeek | 5M free tokens/month | Google Gemini (1,500 req/day) |
| Largest Context | xAI Grok | 2M token window | OpenAI GPT-4.1 / Claude / Gemini (1M) |
| Fastest Inference | Groq | 1,000+ tokens/sec | Fireworks / Together AI |
| Best Reasoning | Claude Opus 4.6 / o3 | - | DeepSeek R1 (10x cheaper) |
| Best Coding | Claude 3.5 Sonnet / GPT-4.1 | - | DeepSeek V3 |
| Best Multimodal | Gemini 2.5 Pro | Images+audio+video | GPT-4o |
| Best EU/Privacy | Mistral AI | EU data residency | Google Gemini EU |
| Best Embeddings | Cohere Embed 4 | $0.12/M | OpenAI text-embedding-3-small |
| Best Self-Hosted | Meta Llama 4 | Free open weights | Mistral (open-weight) |
SDK: pip install openai / npm install openai
| Model | Input $/1M | Output $/1M | Context | Best For |
|---|---|---|---|---|
gpt-4.1 |
$2.00 | $8.00 | 1M | Best overall coding + reasoning |
gpt-4.1-mini |
$0.40 | $1.60 | 1M | Balanced cost/quality |
gpt-4.1-nano |
$0.10 | $0.40 | 1M | High-volume, simple tasks |
gpt-4o |
$2.50 | $10.00 | 128K | Multimodal (images, audio) |
gpt-4o-mini |
$0.15 | $0.60 | 128K | Fast + cheap, most popular |
o3 |
$2.00 | $8.00 | 200K | Extended reasoning chains |
o4-mini |
$1.10 | $4.40 | 200K | Fast reasoning, cheap |
from openai import OpenAI
client = OpenAI(api_key="sk-...")
r = client.chat.completions.create(model="gpt-4o-mini", messages=[{"role":"user","content":"Hello"}])
print(r.choices[0].message.content)
SDK: pip install anthropic / npm install @anthropic-ai/sdk
| Model | Input $/1M | Output $/1M | Context | Best For |
|---|---|---|---|---|
claude-opus-4-6 |
$5.00 | $25.00 | 1M | Hardest reasoning, research |
claude-sonnet-3-7 |
$3.00 | $15.00 | 1M | Best agent / tool use |
claude-sonnet-3-5 |
$3.00 | $15.00 | 200K | Best coding, daily driver |
claude-haiku-4-5 |
$1.00 | $5.00 | 1M | Fast + affordable |
SDK: pip install google-generativeai
| Model | Input $/1M | Output $/1M | Context | Best For |
|---|---|---|---|---|
gemini-2.5-pro |
$1.25 | $10.00 | 1M | Best multimodal, 1M context |
gemini-2.5-flash |
$0.15 | $0.60 | 1M | Speed + budget balance |
gemini-2.5-flash-lite |
$0.10 | $0.40 | 1M | Cheapest capable model |
Free Tier (no credit card): 1,500 requests/day on Flash models
SDK: OpenAI-compatible (pip install openai), Base URL: https://api.deepseek.com/v1
| Model | Input $/1M | Output $/1M | Context | Best For |
|---|---|---|---|---|
deepseek-chat (V3) |
$0.28 | $1.10 | 128K | Best value for money -- GPT-4 quality at 1/10 cost |
deepseek-reasoner (R1) |
$0.55 | $2.19 | 128K | Reasoning chains, math, logic |
Free tier: 5M tokens/month. No credit card needed.
SDK: OpenAI-compatible, Base URL: https://api.x.ai/v1
| Model | Input $/1M | Output $/1M | Context | Best For |
|---|---|---|---|---|
grok-4 |
$3.00 | $15.00 | 2M | Largest public context window |
grok-4.1 |
$2.00 | $10.00 | 2M | Long document analysis |
grok-4.1-fast |
$0.20 | $0.80 | 128K | Fast, cheap Grok |
SDK: OpenAI-compatible (pip install groq), Base URL: https://api.groq.com/openai/v1
| Model | Input $/1M | Output $/1M | Speed | Context |
|---|---|---|---|---|
llama-4-maverick |
$0.50 | $0.77 | 800 tok/s | 128K |
llama-4-scout |
$0.11 | $0.34 | 1,000+ tok/s | 128K |
llama-3.1-8b |
$0.05 | $0.08 | 1,200 tok/s | 128K |
whisper-large-v3 |
$0.111/hr | - | Fastest STT | Audio |
Real-time use case: Groq is the go-to for human-speed responses -- voice apps, real-time agents, streaming UIs.
SDK: OpenAI-compatible, Base URL: https://openrouter.ai/api/v1
Route to any model from a single API key. Auto-fallback, load balancing, unified billing.
| Category | Top Models Available |
|---|---|
| Top Tier | Claude Opus, GPT-4o, Gemini 2.5 Pro, Grok 4 |
| Mid Tier | Claude Sonnet, GPT-4o-mini, Gemini Flash |
| Budget | DeepSeek V3, Llama 4, Mistral Small, Qwen |
| Free | 50+ free models (rate limited) |
| Model | Parameters | VRAM Required | Context |
|---|---|---|---|
Llama-3.2-1B |
1B | 2 GB | 128K |
Llama-3.2-3B |
3B | 4 GB | 128K |
Llama-3.1-8B |
8B | 8 GB | 128K |
Llama-3.3-70B |
70B | 40 GB | 128K |
Llama-4-Scout |
17B active (109B MoE) | 24 GB | 10M |
Llama-4-Maverick |
17B active (400B MoE) | 80 GB | 1M |
Run locally: ollama run llama3.3:70b
Each provider has a dedicated markdown file in /api_providers/:
| File | Contents |
|---|---|
major_cloud_providers/OpenAI.md |
All models, pricing, rate limits, endpoints, code examples |
major_cloud_providers/Anthropic.md |
Claude models, tool use, caching, streaming |
major_cloud_providers/Gemini.md |
All Gemini models, multimodal, free tier details |
major_cloud_providers/DeepSeek.md |
V3 and R1 full breakdown |
major_cloud_providers/Grok.md |
xAI models and 2M context window details |
major_cloud_providers/Mistral.md |
EU-hosted options, codestral, embedding |
major_cloud_providers/Cohere.md |
RAG-specialized, reranking, embed v4 |
major_cloud_providers/Groq.md |
Speed benchmarks, all supported models |
unified_api_platforms/OpenRouter.md |
300+ model catalogue, free tier list |
cloud_provider_ai_services/Azure_OpenAI.md |
Enterprise compliance, PTU pricing |
cloud_provider_ai_services/AWS_Bedrock.md |
IAM setup, on-demand vs provisioned |
open_source_models/Meta_Llama.md |
Download links, hardware requirements |
_COMPARISON.md |
Side-by-side winner table for every category |
Pillar 6 -- No-Code and Visual Workflow Builders
/nocode_platforms/-- 15 fully cloned repositories across 5 categories -- AI workflow builders, low-code internal tools, n8n templates, Flowise tools, and additional platforms
| Folder | Repository | Stars | What It Is |
|---|---|---|---|
n8n |
n8n-io/n8n | 181K+ | Most popular open-source workflow automation. 400+ integrations, native AI, self-hostable |
dify |
langgenius/dify | 135K+ | Full-stack LLMOps platform. Visual workflow builder with RAG, 50+ model providers |
langflow |
langflow-ai/langflow | 146K+ | Drag-and-drop builder for LangChain and RAG applications |
Flowise |
FlowiseAI/Flowise | 51K+ | Visual LangChain + AI agent builder. 100+ integrations |
activepieces |
activepieces/activepieces | 21K+ | AI Agents + MCP + automation. Zapier alternative |
trigger_dev |
triggerdotdev/trigger.dev | 14K+ | Developer-first background jobs. Perfect for AI pipelines |
windmill |
windmill-labs/windmill | 16K+ | 13x faster than Airflow. Retool alternative |
| Folder | Repository | Stars | What It Is |
|---|---|---|---|
appsmith |
appsmithorg/appsmith | 39K+ | Build admin panels, dashboards. 25+ DB integrations |
ToolJet |
ToolJet/ToolJet | 37K+ | AI-native low-code platform. 50+ UI components |
budibase |
[Budibase/budibase](https://github.com/Budibase |
Comments (0)
Sign in to join the discussion.
No comments yet
Be the first to share your take.