LLM Hub — Models, Inference, RAG, LLMOps & Gateways

This hub is the starting point for the LLM material on peterindia.net. Each spoke below is a directory or technical guide of its own, grouped by the stage of the LLM lifecycle it covers.

29Spoke pages

6Topic areas

Models, Assistants & Architectures

Which language models exist, how the architectures differ, and where to find open, compact and assistant models.

01Large Language Models (LLMs)Directory of 43 frontier, open-weight and earlier-generation models and assistants. 02Open-Source LLMsLeading open-weight language models you can download, run and adapt. 03Small Language Models (SLMs)Compact models built for on-device and low-cost use, with how they compare. 04Language ModelsGPT, mixture-of-experts, reasoning, small and action models explained, plus papers to read. 05ChatGPT AlternativesSeventeen AI chatbots, assistants and writing tools to compare.

Building, Training & Learning

How models get built, plus the libraries and learning resources for developers working with them.

06Steps to Build an LLMThe training pipeline from data collection and tokenization to alignment and evaluation. 07Training & Inference OptimizationParallelism, memory budgeting, KV cache and batching for training and serving at scale. 08Python Libraries for LLM ApplicationsTen libraries for model access, orchestration, RAG, fine-tuning and evaluation. 09AI & LLM Learning ResourcesFree courses, guides and repositories for learning LLMs, agents and prompt engineering.

Inference & Serving

Running models efficiently, on a laptop or across a GPU fleet.

10LLM Inference EnginesvLLM, SGLang, TensorRT-LLM, llama.cpp and Ollama compared. 11Platforms for Running LLMs LocallyOllama, LM Studio, llama.cpp and other tools for private, on-device models. 12Inference Optimization TechniquesTen ways to make inference faster and cheaper, from KV caching to quantization. 13Inference Cost-Reduction MethodsHow production teams cut costs with routing, caching, batching and speculative decoding. 14LLM Model Formats ExplainedHow GGUF, GPTQ, AWQ and EXL2 differ and when to use each. 15The Production AI Inference StackvLLM, llm-d and distributed serving, with prefill/decode disaggregation. 16Load Balancing GPU InferenceHow clusters route LLM traffic using KV-cache-aware and disaggregated serving.

RAG & Retrieval

Grounding models in your own data with retrieval-augmented generation.

17RAG Architectures ElucidatedPipeline stages, common pitfalls and what production readiness takes. 18Trends and Transitions in RAGFrom naive pipelines to agentic retrieval. 19Chunking Strategies for RAGHow splitting documents affects retrieval quality. 20Top Reranking ModelsRerankers that improve the results retrieved for RAG.

LLMOps, Observability & Gateways

Operating LLM applications in production: tooling, tracing and the traffic control layer.

21LLMOpsThe practices for operating LLM applications: a nine-stage lifecycle, checklist and risks. 22LLMOps ToolsTen core tools for tracing, evaluating, guarding and serving LLM applications. 23LLM Observability ToolsThirty-nine platforms for tracing, evaluating and monitoring LLMs and agents. 24LLM GatewaysGateways and proxies that unify model-provider APIs with routing, failover and cost control. 25AI GatewaysPlatforms that govern LLM, MCP and agent traffic with security and observability. 26Top AI GatewaysA short list of leading gateways for routing and monitoring LLM traffic. 27Distinct Capabilities of AI GatewaysTwelve things that separate an AI gateway from a traditional API gateway.

Hallucination & Output Quality

Keeping model output accurate and grounded.

28Hallucination-Mitigation MechanismsA taxonomy of grounding, constraint, verification and adaptation techniques. 29Reducing LLM HallucinationEight production techniques, from retrieval and reranking to verification and gating.