awesome-LLM-resources

Organizes 300+ LLM ecosystem resources across 25+ categories using a bilingual (Chinese/English) hierarchical markdown structure deployed via Jekyll GitHub Pages. The catalog uses a consistent section pattern with category headers, resource links, and descriptions that enable both human browsing and programmatic discovery through GitHub's raw markdown API. Each resource is tagged with domain (foundation, deployment, multimodal, etc.) enabling cross-domain navigation and filtering.

Catalogs 40+ resources spanning data processing, model training, fine-tuning frameworks, and reinforcement learning approaches. The catalog maps the complete pipeline from raw data curation through foundation model training, including tools for data annotation (Label Studio, Argilla), preprocessing (Hugging Face Datasets), fine-tuning (Unsloth, LLaMA-Factory), and agentic RL (veRL, AReaL). Resources are organized by training methodology (supervised fine-tuning, RLHF, DPO, GRPO) enabling builders to select appropriate frameworks for their training objectives.

Catalogs 15+ resources for advanced reasoning models (OpenAI o1, o3, DeepSeek-R1) and open-source reasoning model implementations. The catalog maps how reasoning models differ from standard LLMs (chain-of-thought training, test-time compute, verification), including training approaches (GRPO, RL-based reasoning) and inference patterns. Resources span both commercial reasoning APIs and open-source implementations, enabling builders to understand and implement advanced reasoning capabilities.

Catalogs 25+ small and efficient LLM models (Phi, TinyLlama, Mistral 7B, Qwen, Gemma) organized by optimization approach: quantization (GPTQ, AWQ, GGUF), distillation, pruning, and architectural efficiency. The catalog maps how efficient models trade off capability for size/speed, including benchmarks on standard tasks. Resources span both pre-optimized models and optimization frameworks, enabling builders to select or create efficient models for resource-constrained deployments.

Catalogs resources for Model Context Protocol (MCP), a standardized protocol for LLM context management and tool integration. The catalog maps MCP implementations, client libraries, and server implementations, including integration patterns with LLM applications. Resources span both MCP specification documentation and practical implementations, enabling builders to understand and implement MCP-based context management and tool orchestration.

Catalogs 50+ learning resources organized by format: books (LLM fundamentals, prompt engineering, RAG), courses (university courses, online platforms), and technical papers (foundational research, recent advances). The catalog maps resources by topic (transformer architecture, fine-tuning, agents, multimodal) and audience level (beginner, intermediate, advanced), enabling learners to find appropriate educational materials for their background and goals.

Catalogs 10+ interactive platforms (Hugging Face Spaces, OpenRouter, Chatbot Arena, Together Playground) enabling side-by-side model comparison and evaluation. The catalog maps how platforms enable comparative evaluation (same prompt across models, user voting, leaderboards) and integration with multiple model providers. Resources span both community-driven arenas (Chatbot Arena) and commercial platforms (OpenRouter), enabling builders to evaluate models before integration.

Aggregates 30+ inference serving frameworks (vLLM, TensorRT-LLM, SGLang, Ollama, LM Studio) organized by deployment pattern (local, cloud, edge, batch). The catalog maps frameworks to specific optimization techniques (quantization, batching, KV-cache optimization) and hardware targets (CPU, GPU, mobile). Resources include both open-source inference engines and commercial serving platforms, enabling builders to select frameworks matching their latency, throughput, and cost requirements.

Catalogs 50+ RAG components organized by pipeline stage: document ingestion (LlamaIndex, LangChain), vector databases (Pinecone, Weaviate, Milvus, Qdrant), retrieval optimization (BM25, semantic search, hybrid retrieval), and generation orchestration. The catalog maps how components integrate into end-to-end RAG pipelines, including chunking strategies, embedding models, reranking, and prompt engineering techniques. Resources span both framework-level solutions (LlamaIndex, LangChain) and specialized components (vector databases, rerankers).

Aggregates 40+ agent frameworks (AutoGen, LangGraph, CrewAI, Swarm, Rigging) organized by orchestration pattern: multi-agent coordination, tool calling, memory management, and planning strategies. The catalog maps how frameworks implement agent capabilities (function calling, state management, tool registry) and integration points with LLM APIs (OpenAI, Anthropic, Ollama). Resources include both high-level agent frameworks and lower-level orchestration primitives, enabling builders to select frameworks matching their agent complexity and coordination requirements.

Catalogs 25+ coding-focused LLM tools (GitHub Copilot, Cursor, Codeium, Aider, Continue) organized by capability: code completion, refactoring, debugging, code review, and test generation. The catalog maps tools by integration point (IDE plugins, CLI, web-based) and supported languages/frameworks. Resources include both commercial coding assistants and open-source alternatives, enabling developers to select tools matching their development workflow and language preferences.

Aggregates 20+ search and research tools (Perplexity, Tavily, Exa, Metaphor, web search APIs) organized by retrieval pattern: web search, academic paper search, semantic search, and real-time information retrieval. The catalog maps how tools integrate with LLM applications through APIs, including search result formatting and citation handling. Resources span both consumer-facing search tools and developer-oriented search APIs, enabling builders to select tools matching their information retrieval requirements.

Catalogs 60+ multimodal resources organized by modality: vision-language models (GPT-4V, Claude Vision, LLaVA), video generation (Sora, Runway, Pika), image generation (DALL-E, Midjourney, Stable Diffusion), speech systems (Whisper, TTS, voice cloning), and unified multimodal models (Gemini, GPT-4o). The catalog maps how multimodal models integrate with LLM applications, including input/output format handling and API integration patterns. Resources span both commercial APIs and open-source models, enabling builders to select tools matching their multimodal requirements.

Aggregates 30+ evaluation frameworks and benchmarks organized by evaluation type: LLM capability benchmarks (MMLU, HumanEval, MATH), RAG evaluation (RAGAS, TruLens), agent evaluation (AgentBench), and safety/alignment evaluation. The catalog maps how evaluation frameworks measure specific capabilities (reasoning, coding, knowledge, safety) and integrate with model development pipelines. Resources span both standardized benchmarks (MMLU, HumanEval) and specialized evaluation tools (RAGAS for RAG, TruLens for observability).

Catalogs 15+ LLM API providers (OpenAI, Anthropic, Google, Meta, Mistral, Qwen, Together AI, Replicate) organized by provider type: frontier models (GPT-4, Claude), open-source model APIs (Mistral, Qwen, Llama), and specialized providers (Together AI for fine-tuning, Replicate for inference). The catalog maps API capabilities (model versions, context length, pricing, rate limits) and integration patterns, enabling builders to select providers matching their cost, latency, and capability requirements.