Persistent Llm Memory

Pieces stores and organizes work‑related context—code, docs, chats—within familiar tools, creating OS‑level long‑term memory. It supports real‑time LLM context via local plugins, letting users keep data on‑device or sync to a chosen cloud, aiding continuity for teams.

LiteLLM is an open‑source gateway that unifies access to 100+ LLMs through a single OpenAI‑compatible API, enabling provider fallback, cost tracking, tag‑based budgeting, guardrails, observability, and on‑prem or cloud deployment with a lightweight SDK.

llongterm is a memory-focused AI tool that enhances chatbots and virtual assistants by enabling contextual memory retention. It supports personalized interactions across various applications, including education and customer support, and is compatible with multiple programming languages.

Awan LLM offers unlimited token generation with Meta Llama 3.1 8B and 70B models, no censorship or caps, supporting persistent AI assistance, autonomous agents, roleplay, data processing, and code completion, hosted on owned GPUs for continuous use.

Remind AI is an open‑source memory system that records and indexes digital activity. It stores data from email, messaging, and document editors in a structured graph, enabling users to retrieve past actions and files with natural language via local LLMs.

LLMStack is an open‑source platform that lets developers build AI agents and workflows without coding, supports multiple model providers, imports data from web, PDFs, audio, cloud services, and offers a collaborative React UI with granular permissions.

LLM Pulse tracks brand visibility and search presence across LLMs (ChatGPT, Perplexity, Google AI), offering prompt tracking and suggestions, citation analysis, visibility scoring and competitor benchmarking, sentiment and response inspection, plus API and reporting exports.

LLMChat is an AI chat tool that offers a beta version experience with diverse AI models, personalized memory, custom assistant creation, and privacy-focused locally stored conversations. Explore features like plugin integration, tailored preferences, and prompt examples for various tasks.

RunLLM is an AI platform that automates incident investigations by querying observability tools, correlating telemetry, and delivering root-cause analyses. It generates live runbooks and remediation recommendations to accelerate MTTR and create an auditable history of incidents.

Inception Labs' diffusion-based large language models (dLLMs) offer faster, more efficient, and cost-effective text generation than traditional autoregressive models. With built-in error correction, multimodal support, and structured output control, they excel in function calling and complex data ge

VLLM is a high-throughput, memory-efficient inference engine for Large Language Models, enabling faster responses and effective memory management. It supports multi-node configurations for scalability and offers robust documentation for seamless integration into workflows.

LLMWare AI installs a lightweight client on PCs, providing instant access to 100+ AI models optimized for Intel and Qualcomm hardware. It supports RAG, auto‑tunes weights, runs locally without Wi‑Fi, and offers an admin console for monitoring, scaling, and audit logs.

LastMile AI is a platform that perceives, remembers, and reasons from vision, speech, and text using LLMs as CPU and context as RAM. It connects to tools, automates workflows, anticipates needs, and surfaces actionable insights for teams and organizations.

Reflection 70B is an open‑source 70 B Llama 3.1‑based model that uses real‑time reflection tuning for self‑correction. It outperforms GPT‑4o on MMLU, HumanEval, MATH, IFEval, GSM8K, supporting accurate coding, debugging, and reasoning tasks via API, with a no‑registration web interface.

Pocketllm is an AI-powered personal document search engine that allows you to easily search and retrieve information from thousands of pages of PDFs and documents. It offers semantic search capability, fine-tuning search results and summarizing results.

exllama is a memory-efficient tool for executing Hugging Face transformers with the LLaMA models using quantized weights, enabling high-performance NLP tasks on modern GPUs while minimizing memory usage and supporting various hardware configurations.

Micro LLM is a personal AI assistant that enhances productivity by managing tasks, scheduling appointments, and answering questions. It operates on devices like iPads and iPhones, offering offline functionality and an intuitive interface for seamless organization.

Mistral.rs is an efficient, versatile tool for high-speed large language model (LLM) inference, offering multi-device support and extensive quantization options for seamless deployment on diverse hardware setups.

EverMemOS is a long-term memory operating system for AI agents that converts interactions into structured MemUnits, organizes hierarchical adaptive memory graphs, and provides indexed retrieval, reranking, and APIs to enable persistent, context-aware, stateful behavior across sessions.

little-coder is a Pi-based coding agent for running 5–25 GB local LLMs via llama.cpp or Ollama, offering Python/Node CLIs and TypeScript extensions, reproducible benchmarks, build/serve guides, and tools for local code generation, on-device development, and evaluation.