https://ramdi.fr/tags/llm/Recent content in Llm on Noureddine RAMDIHugoenSat, 23 May 2026 20:41:27 +0000https://ramdi.fr/github-stars/ai-interview-codex-iterative-ai-system-design-and-interview-prep-with-real-world-benchmarks/Sat, 23 May 2026 20:41:14 +0000https://ramdi.fr/github-stars/ai-interview-codex-iterative-ai-system-design-and-interview-prep-with-real-world-benchmarks/ai-interview-codex offers a practical AI interview prep guide featuring iterative system design for Agentic AI and RAG, with benchmarks and production insights for ML, LLM, and system design roles.https://ramdi.fr/github-stars/arkon-structured-enterprise-knowledge-synthesis-with-a-unique-llm-compilation-pipeline/Sat, 23 May 2026 20:41:14 +0000https://ramdi.fr/github-stars/arkon-structured-enterprise-knowledge-synthesis-with-a-unique-llm-compilation-pipeline/Arkon is a self-hosted enterprise knowledge hub using a novel MRP pipeline for structured, traceable wiki compilation with external AI inference and workspace-scoped RBAC.https://ramdi.fr/github-stars/autoskill-experience-driven-lifelong-learning-for-llm-agents-with-skill-versioning-and-evolution/Sat, 23 May 2026 20:41:14 +0000https://ramdi.fr/github-stars/autoskill-experience-driven-lifelong-learning-for-llm-agents-with-skill-versioning-and-evolution/AutoSkill is a Python framework enabling LLM agents to extract, version, and evolve skills from dialogues, providing a persistent long-term memory system for AI agents.https://ramdi.fr/github-stars/chattutor-enabling-ai-tutors-to-teach-stem-visually-with-a-vue-bun-full-stack/Sat, 23 May 2026 20:41:14 +0000https://ramdi.fr/github-stars/chattutor-enabling-ai-tutors-to-teach-stem-visually-with-a-vue-bun-full-stack/ChatTutor integrates AI tutors with visual tools like Geogebra in a Vue + Bun full-stack. It supports multiple LLM providers and offers a digital whiteboard for interactive STEM learning.https://ramdi.fr/github-stars/claw-eval-a-rigorous-python-harness-for-trustworthy-evaluation-of-llm-powered-autonomous-agents/Sat, 23 May 2026 20:41:14 +0000https://ramdi.fr/github-stars/claw-eval-a-rigorous-python-harness-for-trustworthy-evaluation-of-llm-powered-autonomous-agents/Claw-Eval offers a Python-based evaluation harness for LLM autonomous agents, featuring 300 tasks and a strict Pass^3 metric to ensure reliable, multi-dimensional benchmarking.https://ramdi.fr/github-stars/comic-translate-ai-driven-multi-language-comic-translation-with-full-page-context/Sat, 23 May 2026 20:41:14 +0000https://ramdi.fr/github-stars/comic-translate-ai-driven-multi-language-comic-translation-with-full-page-context/Comic Translate uses advanced AI models and a multi-step pipeline for accurate comic translation across languages, combining speech bubble detection, OCR, and LLMs with full-page context.https://ramdi.fr/github-stars/deepteam-a-python-framework-for-adversarial-red-teaming-of-large-language-models/Sat, 23 May 2026 20:41:14 +0000https://ramdi.fr/github-stars/deepteam-a-python-framework-for-adversarial-red-teaming-of-large-language-models/DeepTeam is a Python tool for red teaming LLMs by dynamically generating adversarial attacks and evaluating vulnerabilities like bias. It requires minimal setup and no predefined datasets.https://ramdi.fr/github-stars/deepzero-automating-windows-kernel-driver-vulnerability-research-with-yaml-driven-llm-pipelines/Sat, 23 May 2026 20:41:14 +0000https://ramdi.fr/github-stars/deepzero-automating-windows-kernel-driver-vulnerability-research-with-yaml-driven-llm-pipelines/DeepZero automates vulnerability research on Windows kernel drivers by chaining Ghidra decompilation with LLM-based analysis using YAML pipelines and Jinja2 templates.https://ramdi.fr/github-stars/dot-an-offline-electron-desktop-app-for-local-llm-inference-and-document-qa/Sat, 23 May 2026 20:41:14 +0000https://ramdi.fr/github-stars/dot-an-offline-electron-desktop-app-for-local-llm-inference-and-document-qa/Dot bundles local LLM inference, Retrieval Augmented Generation, and Text-To-Speech into a single offline Electron app, enabling document QA without cloud dependencies.https://ramdi.fr/github-stars/fuzzyai-ai-driven-fuzz-testing-with-local-llm-integration/Sat, 23 May 2026 20:41:14 +0000https://ramdi.fr/github-stars/fuzzyai-ai-driven-fuzz-testing-with-local-llm-integration/FuzzyAI combines fuzz testing with AI models using Python and Ollama. It offers a CLI for fuzzing with local LLMs, balancing AI power and practical setup tradeoffs.https://ramdi.fr/github-stars/harvey-lab-benchmarking-legal-llm-agents-with-realistic-tasks-and-automated-scoring/Sat, 23 May 2026 20:41:14 +0000https://ramdi.fr/github-stars/harvey-lab-benchmarking-legal-llm-agents-with-realistic-tasks-and-automated-scoring/Harvey LAB offers an open-source benchmark for evaluating LLM agents on realistic legal tasks using an all-pass rubric and LLM-as-judge scoring. It includes datasets, adapters, and dashboards.https://ramdi.fr/github-stars/inside-mini-sglang-a-clear-and-modular-python-llm-inference-engine/Sat, 23 May 2026 20:41:14 +0000https://ramdi.fr/github-stars/inside-mini-sglang-a-clear-and-modular-python-llm-inference-engine/Mini-SGLang is a modular Python reimplementation of the SGLang LLM inference engine with production features like Radix Cache, chunked prefill, overlap scheduling, and tensor parallelism.https://ramdi.fr/github-stars/inside-picoagents-a-transparent-multi-agent-system-framework-built-from-scratch-in-python/Sat, 23 May 2026 20:41:14 +0000https://ramdi.fr/github-stars/inside-picoagents-a-transparent-multi-agent-system-framework-built-from-scratch-in-python/PicoAgents is a Python multi-agent framework built from scratch, offering transparent agent orchestration, LLM provider abstraction, streaming UI, and production-ready benchmarks.https://ramdi.fr/github-stars/inside-xalgorix-an-llm-driven-autonomous-pentesting-platform-with-a-22-phase-testing-pipeline/Sat, 23 May 2026 20:41:14 +0000https://ramdi.fr/github-stars/inside-xalgorix-an-llm-driven-autonomous-pentesting-platform-with-a-22-phase-testing-pipeline/Xalgorix is a Go-based autonomous pentesting platform driven by LLMs, featuring a 22-phase methodology from recon to exploit verification, with live telemetry and reporting.https://ramdi.fr/github-stars/intellagent-systematic-adversarial-testing-for-conversational-ai-with-policy-graph-decomposition/Sat, 23 May 2026 20:41:14 +0000https://ramdi.fr/github-stars/intellagent-systematic-adversarial-testing-for-conversational-ai-with-policy-graph-decomposition/IntellAgent is a Python framework that stress-tests conversational AI agents by generating structured adversarial dialogues via policy graph decomposition, helping uncover blind spots before production.https://ramdi.fr/github-stars/kimi-audio-a-unified-hybrid-token-audio-foundation-model-with-llm-core/Sat, 23 May 2026 20:41:14 +0000https://ramdi.fr/github-stars/kimi-audio-a-unified-hybrid-token-audio-foundation-model-with-llm-core/Kimi-Audio combines continuous acoustic and discrete semantic tokens within a 7B LLM for unified audio-text understanding and generation. It achieves state-of-the-art ASR with low-latency audio synthesis.https://ramdi.fr/github-stars/livecaptions-translator-real-time-speech-translation-using-windows-11-s-built-in-captions-and-llm-apis/Sat, 23 May 2026 20:41:14 +0000https://ramdi.fr/github-stars/livecaptions-translator-real-time-speech-translation-using-windows-11-s-built-in-captions-and-llm-apis/LiveCaptions Translator taps Windows 11’s on-device LiveCaptions for real-time speech translation via multiple LLM and traditional APIs, all in a sleek C# desktop app.https://ramdi.fr/github-stars/livetradebench-evaluating-llm-driven-trading-agents-in-live-markets/Sat, 23 May 2026 20:41:14 +0000https://ramdi.fr/github-stars/livetradebench-evaluating-llm-driven-trading-agents-in-live-markets/LiveTradeBench benchmarks LLM trading agents like GPT and Claude in live US equity and prediction markets with real-time news and sentiment integration.https://ramdi.fr/github-stars/llm-mm-agent-autonomous-mathematical-modeling-with-hierarchical-method-selection/Sat, 23 May 2026 20:41:14 +0000https://ramdi.fr/github-stars/llm-mm-agent-autonomous-mathematical-modeling-with-hierarchical-method-selection/LLM-MM-Agent uses LLMs as autonomous agents for end-to-end mathematical modeling, featuring a unique hierarchical method library with actor-critic selection. Supports GPT-4o and DeepSeek-R1.https://ramdi.fr/github-stars/llm4pentest-a-curated-knowledge-hub-on-large-language-models-for-automated-penetration-testing/Sat, 23 May 2026 20:41:14 +0000https://ramdi.fr/github-stars/llm4pentest-a-curated-knowledge-hub-on-large-language-models-for-automated-penetration-testing/LLM4Pentest aggregates 40+ research papers and tools tracking the evolving role of LLMs in automated penetration testing, highlighting progress and limitations.https://ramdi.fr/github-stars/llmstxt-architect-automated-generation-and-maintenance-of-llms-txt-files-for-llm-aware-websites/Sat, 23 May 2026 20:41:14 +0000https://ramdi.fr/github-stars/llmstxt-architect-automated-generation-and-maintenance-of-llms-txt-files-for-llm-aware-websites/llmstxt_architect automates generating and updating llms.txt files that communicate website content to LLMs. Supports multi-provider LLMs and preserves file structure during updates.https://ramdi.fr/github-stars/macai-a-unified-native-macos-ai-chat-client-for-multiple-llm-providers/Sat, 23 May 2026 20:41:14 +0000https://ramdi.fr/github-stars/macai-a-unified-native-macos-ai-chat-client-for-multiple-llm-providers/macai is a native macOS AI chat client unifying access to major LLM providers with iCloud Sync and local inference support, offering a minimalist cross-device AI chat experience.https://ramdi.fr/github-stars/markpdfdown-converting-pdfs-to-markdown-using-vision-capable-large-language-models/Sat, 23 May 2026 20:41:14 +0000https://ramdi.fr/github-stars/markpdfdown-converting-pdfs-to-markdown-using-vision-capable-large-language-models/MarkPDFDown is a Python CLI tool that converts PDFs and images into Markdown by using vision-capable large language models for visual recognition-based parsing, handling complex layouts and formulas.https://ramdi.fr/github-stars/minimalist-python-ai-demos-exploring-qxresearch-event-1-s-concise-llm-patterns/Sat, 23 May 2026 20:41:14 +0000https://ramdi.fr/github-stars/minimalist-python-ai-demos-exploring-qxresearch-event-1-s-concise-llm-patterns/qxresearch-event-1 is a collection of 50+ minimalist Python apps showcasing core AI patterns like fine-tuning, vector DB, and Whisper in about 10 lines each. A practical learning resource.https://ramdi.fr/github-stars/navigating-the-evolving-landscape-of-llm-based-multi-agent-systems-a-survey-paper-repository/Sat, 23 May 2026 20:41:14 +0000https://ramdi.fr/github-stars/navigating-the-evolving-landscape-of-llm-based-multi-agent-systems-a-survey-paper-repository/A curated and frequently updated bibliography accompanying the IJCAI 2024 survey paper on LLM-based multi-agent systems, organizing research into five key categories and revealing emerging trends.https://ramdi.fr/github-stars/navigating-the-llm-engineer-handbook-a-curated-map-for-production-grade-language-models/Sat, 23 May 2026 20:41:14 +0000https://ramdi.fr/github-stars/navigating-the-llm-engineer-handbook-a-curated-map-for-production-grade-language-models/The LLM Engineer Handbook catalogs the full lifecycle of large language model engineering, from pretraining to prompt management, guiding engineers beyond demos to production-ready LLM apps.https://ramdi.fr/github-stars/nomai-a-multi-step-ai-nutrition-analysis-app-combining-flutter-and-fastapi/Sat, 23 May 2026 20:41:14 +0000https://ramdi.fr/github-stars/nomai-a-multi-step-ai-nutrition-analysis-app-combining-flutter-and-fastapi/NomAI combines a Flutter app with a FastAPI backend using a multi-step LLM pipeline and web-grounded reasoning for nutrition analysis and meal tracking.https://ramdi.fr/github-stars/npcpy-enforcing-ai-behavioral-compliance-through-architecture-for-multimodal-llm-apps/Sat, 23 May 2026 20:41:14 +0000https://ramdi.fr/github-stars/npcpy-enforcing-ai-behavioral-compliance-through-architecture-for-multimodal-llm-apps/npcpy offers a unique NPC Context-Agent-Tool data layer to enforce AI compliance via software architecture, supporting multimodal LLM apps and multi-agent systems with local and cloud providers.https://ramdi.fr/github-stars/obsidian-llm-wiki-local-local-first-ai-powered-wiki-generation-with-human-in-the-loop-feedback/Sat, 23 May 2026 20:41:14 +0000https://ramdi.fr/github-stars/obsidian-llm-wiki-local-local-first-ai-powered-wiki-generation-with-human-in-the-loop-feedback/obsidian-llm-wiki-local generates interlinked Obsidian markdown wikis using local LLMs. Its standout feature is a rejection feedback loop that refines article quality via user input.https://ramdi.fr/github-stars/open-computer-use-orchestrating-multi-model-llm-pipelines-for-remote-linux-desktop-control/Sat, 23 May 2026 20:41:14 +0000https://ramdi.fr/github-stars/open-computer-use-orchestrating-multi-model-llm-pipelines-for-remote-linux-desktop-control/Open Computer Use uses a modular three-stage LLM pipeline to control a cloud Linux desktop, combining grounding, vision, and action models for flexible AI-driven automation.https://ramdi.fr/github-stars/openagents-orchestrating-multi-agent-llm-workflows-with-flask-and-next-js/Sat, 23 May 2026 20:41:14 +0000https://ramdi.fr/github-stars/openagents-orchestrating-multi-agent-llm-workflows-with-flask-and-next-js/OpenAgents hosts three specialized LLM agents—Data, Plugins, Web—via a Flask API and Next.js UI, integrating sandboxed code execution, plugin selection, and browser automation.https://ramdi.fr/github-stars/openant-an-llm-powered-two-stage-vulnerability-discovery-tool-with-exploit-validation/Sat, 23 May 2026 20:41:14 +0000https://ramdi.fr/github-stars/openant-an-llm-powered-two-stage-vulnerability-discovery-tool-with-exploit-validation/OpenAnt uses a two-stage LLM pipeline to detect and validate code vulnerabilities across multiple languages, reducing false positives by verifying exploits automatically.https://ramdi.fr/github-stars/openchronicle-an-ax-first-local-memory-layer-for-llm-agents/Sat, 23 May 2026 20:41:14 +0000https://ramdi.fr/github-stars/openchronicle-an-ax-first-local-memory-layer-for-llm-agents/OpenChronicle captures macOS accessibility events to build structured local memory for LLM agents. Its async pipeline produces persistent Markdown memory and an SQLite index.https://ramdi.fr/github-stars/optillm-transparent-inference-time-scaling-for-improved-llm-reasoning/Sat, 23 May 2026 20:41:14 +0000https://ramdi.fr/github-stars/optillm-transparent-inference-time-scaling-for-improved-llm-reasoning/OptiLLM is an OpenAI-compatible inference proxy that boosts LLM reasoning with 20+ techniques like Mixture of Agents and MCTS, requiring no model retraining. Use a simple prefix to improve accuracy 2-10x.https://ramdi.fr/github-stars/paper2any-multi-modal-ai-pipeline-converting-academic-papers-into-editable-scientific-artifacts/Sat, 23 May 2026 20:41:14 +0000https://ramdi.fr/github-stars/paper2any-multi-modal-ai-pipeline-converting-academic-papers-into-editable-scientific-artifacts/Paper2Any uses chained LLM calls with structured output to convert academic papers into editable scientific figures, slides, and diagrams via a FastAPI backend and React frontend.https://ramdi.fr/github-stars/reasoningbank-experience-driven-memory-as-a-new-scaling-dimension-for-ai-agents/Sat, 23 May 2026 20:41:14 +0000https://ramdi.fr/github-stars/reasoningbank-experience-driven-memory-as-a-new-scaling-dimension-for-ai-agents/ReasoningBank introduces memory-aware test-time scaling for AI agents by storing reasoning traces from both successes and failures, enabling self-evolution through experience.https://ramdi.fr/github-stars/superclaude-meta-programming-claude-code-into-a-structured-ai-development-platform/Sat, 23 May 2026 20:41:14 +0000https://ramdi.fr/github-stars/superclaude-meta-programming-claude-code-into-a-structured-ai-development-platform/SuperClaude transforms Claude Code into a structured AI development platform using behavioral instruction injection, 30 slash commands, 20 specialized agents, and 8 MCP server integrations for faster, token-efficient workflows.https://ramdi.fr/github-stars/supoclip-self-hostable-ai-powered-video-clipping-with-multi-llm-backend-abstraction/Sat, 23 May 2026 20:41:14 +0000https://ramdi.fr/github-stars/supoclip-self-hostable-ai-powered-video-clipping-with-multi-llm-backend-abstraction/SupoClip is an open-source self-hosted AI video clipper using AssemblyAI transcription and multiple LLM backends including local Ollama. It runs on Docker Compose with FastAPI and Next.js.https://ramdi.fr/github-stars/swark-generating-architecture-diagrams-from-code-using-github-copilot-in-vs-code/Sat, 23 May 2026 20:41:14 +0000https://ramdi.fr/github-stars/swark-generating-architecture-diagrams-from-code-using-github-copilot-in-vs-code/Swark is a VS Code extension that creates Mermaid.js architecture diagrams from any code using GitHub Copilot’s free tier via the VS Code Language Model API—no API keys needed.https://ramdi.fr/github-stars/vllm-compressor-practical-quantization-and-compression-for-large-language-model-inference/Sat, 23 May 2026 20:41:14 +0000https://ramdi.fr/github-stars/vllm-compressor-practical-quantization-and-compression-for-large-language-model-inference/vLLM Compressor applies advanced quantization and compression techniques to large language models, enabling optimized inference without requiring full model definitions.https://ramdi.fr/github-stars/weave-a-go-microkernel-platform-for-hot-pluggable-ai-application-development/Sat, 23 May 2026 20:41:14 +0000https://ramdi.fr/github-stars/weave-a-go-microkernel-platform-for-hot-pluggable-ai-application-development/Weave is a Go-based AI platform with a microkernel architecture that supports hot-pluggable AI plugins and dynamic multi-model switching. Deploy with Docker Compose for rapid development.https://ramdi.fr/github-stars/yoagent-a-minimal-rust-ai-agent-runtime-with-multi-provider-llm-support/Sat, 23 May 2026 20:41:14 +0000https://ramdi.fr/github-stars/yoagent-a-minimal-rust-ai-agent-runtime-with-multi-provider-llm-support/yoagent is a minimal Rust library implementing an AI agent loop with multi-provider LLM support, built-in tools, and event streaming for clean, extensible agent workflows.https://ramdi.fr/github-stars/aider-precise-ai-pair-programming-with-whole-codebase-awareness/Sat, 09 May 2026 11:42:26 +0000https://ramdi.fr/github-stars/aider-precise-ai-pair-programming-with-whole-codebase-awareness/Aider is a terminal-based AI pair programming tool that builds a repository map for full codebase context, enabling precise, developer-controlled edits with multi-LLM support and git integration.https://ramdi.fr/github-stars/inside-claude-code-from-scratch-a-practical-reconstruction-of-anthropic-s-coding-agent/Sat, 09 May 2026 11:42:26 +0000https://ramdi.fr/github-stars/inside-claude-code-from-scratch-a-practical-reconstruction-of-anthropic-s-coding-agent/Claude Code From Scratch distills Anthropic’s 500K+ line coding agent into ~8,000 lines of Python and TypeScript, revealing core architecture like the Agent Loop, semantic memory, multi-agent skills, and context compression.https://ramdi.fr/github-stars/openalpha-evolve-autonomous-code-evolution-with-llm-driven-diff-mutations/Sat, 09 May 2026 11:42:26 +0000https://ramdi.fr/github-stars/openalpha-evolve-autonomous-code-evolution-with-llm-driven-diff-mutations/OpenAlpha_Evolve uses large language models to generate precise code diffs as mutations in an evolutionary algorithm, enabling autonomous iterative code improvement with sandboxed evaluation.https://ramdi.fr/github-stars/ragflow-a-modular-agentic-retrieval-augmented-generation-engine-with-deep-document-understanding/Wed, 06 May 2026 18:58:37 +0000https://ramdi.fr/github-stars/ragflow-a-modular-agentic-retrieval-augmented-generation-engine-with-deep-document-understanding/RAGFlow is an open-source Python RAG engine combining deep document parsing, configurable pipelines, agentic workflows, and sandboxed code execution for LLM context management.https://ramdi.fr/github-stars/rtk-a-rust-cli-proxy-that-cuts-llm-token-usage-by-up-to-90-with-transparent-command-rewriting/Wed, 06 May 2026 18:58:37 +0000https://ramdi.fr/github-stars/rtk-a-rust-cli-proxy-that-cuts-llm-token-usage-by-up-to-90-with-transparent-command-rewriting/rtk is a Rust CLI proxy that intercepts shell commands to reduce LLM token consumption by 60-90% using a transparent Bash hook and output filtering, supporting 100+ commands.https://ramdi.fr/github-stars/10x-cli-coding-agent-tiered-ai-model-routing-for-faster-coding-workflows/Tue, 05 May 2026 22:24:55 +0000https://ramdi.fr/github-stars/10x-cli-coding-agent-tiered-ai-model-routing-for-faster-coding-workflows/10x is a TypeScript CLI coding agent that speeds coding up to 20x by routing tasks across a tiered AI model system with customizable multi-step workflows called Superpowers.https://ramdi.fr/github-stars/langchain-chatchat-a-model-agnostic-orchestration-layer-for-chinese-language-rag-and-agents/Tue, 05 May 2026 22:24:55 +0000https://ramdi.fr/github-stars/langchain-chatchat-a-model-agnostic-orchestration-layer-for-chinese-language-rag-and-agents/Langchain-Chatchat offers a flexible, offline-capable orchestration layer for multiple Chinese LLMs and RAG approaches, enabling seamless model swaps across frameworks without code changes.https://ramdi.fr/github-stars/langfuse-simplifying-llm-observability-with-decorator-based-tracing/Tue, 05 May 2026 22:24:55 +0000https://ramdi.fr/github-stars/langfuse-simplifying-llm-observability-with-decorator-based-tracing/Langfuse provides end-to-end observability for LLM applications with automatic tracing via an @observe() decorator, enabling teams to debug and manage AI workflows efficiently.https://ramdi.fr/github-stars/open-cowork-desktop-ai-agent-with-vm-level-sandbox-isolation-for-safer-ai-workflows/Tue, 05 May 2026 22:24:55 +0000https://ramdi.fr/github-stars/open-cowork-desktop-ai-agent-with-vm-level-sandbox-isolation-for-safer-ai-workflows/Open Cowork wraps multiple LLMs in a cross-platform desktop app with unique VM-level sandboxing using WSL2 and Lima for safe AI agent command execution.https://ramdi.fr/github-stars/quivr-a-python-framework-for-flexible-retrieval-augmented-generation-pipelines/Tue, 05 May 2026 22:24:55 +0000https://ramdi.fr/github-stars/quivr-a-python-framework-for-flexible-retrieval-augmented-generation-pipelines/Quivr is a Python framework offering an opinionated, pluggable retrieval-augmented generation pipeline with multi-LLM support and YAML-defined workflows for flexible knowledge retrieval.https://ramdi.fr/github-stars/inside-genie-sim-3-0-llm-driven-embodied-ai-simulation-with-high-fidelity-3d-scenes/Tue, 05 May 2026 16:46:42 +0000https://ramdi.fr/github-stars/inside-genie-sim-3-0-llm-driven-embodied-ai-simulation-with-high-fidelity-3d-scenes/Genie Sim 3.0 is an open-source platform combining 3D Gaussian Splatting and LLM-driven scene generation for embodied AI simulation, offering large-scale synthetic data and low sim-to-real discrepancy.https://ramdi.fr/github-stars/open-deep-research-a-next-js-16-agentic-ai-assistant-for-iterative-web-research/Tue, 05 May 2026 16:46:42 +0000https://ramdi.fr/github-stars/open-deep-research-a-next-js-16-agentic-ai-assistant-for-iterative-web-research/Open Deep Research is a TypeScript Next.js 16 app that uses an LLM to plan, execute, and iterate web research via Exa and Upstash QStash, producing sourced reports with images.https://ramdi.fr/github-stars/action100m-hierarchical-tree-of-captions-for-multi-scale-video-understanding/Tue, 05 May 2026 13:37:39 +0000https://ramdi.fr/github-stars/action100m-hierarchical-tree-of-captions-for-multi-scale-video-understanding/Action100M provides a hierarchical Tree-of-Captions annotation for 100M video segments, enabling multi-scale video understanding with LLM-generated captions. Explore its structure, tech strengths, and how to access the data.https://ramdi.fr/github-stars/alibaba-s-qwen3-6-efficient-large-scale-llms-with-gated-delta-networks-and-sparse-moe/Tue, 05 May 2026 13:37:39 +0000https://ramdi.fr/github-stars/alibaba-s-qwen3-6-efficient-large-scale-llms-with-gated-delta-networks-and-sparse-moe/Qwen3.6 from Alibaba uses gated delta networks and sparse Mixture-of-Experts to achieve near-397B parameter model performance with only 3B active parameters, supporting 201 languages and 262k context length.https://ramdi.fr/github-stars/automate-a-local-first-ai-hub-exposing-40-tools-via-mcp-over-http/Tue, 05 May 2026 13:37:39 +0000https://ramdi.fr/github-stars/automate-a-local-first-ai-hub-exposing-40-tools-via-mcp-over-http/autoMate exposes 40+ AI tools and 31 SaaS APIs via MCP-over-HTTP on localhost, with encrypted storage and multi-provider LLM support. A local AI infrastructure hub with privacy-first design.https://ramdi.fr/github-stars/building-production-ready-rag-workflows-with-n8n-using-free-json-templates/Tue, 05 May 2026 13:37:39 +0000https://ramdi.fr/github-stars/building-production-ready-rag-workflows-with-n8n-using-free-json-templates/Explore over 200 pre-built n8n workflow templates integrating vector databases, embedding models, and LLMs for rapid RAG workflow prototyping and deployment without coding.https://ramdi.fr/github-stars/clawsync-a-convex-based-multi-agent-ai-platform-with-shared-soul-documents-and-per-agent-model-routing/Tue, 05 May 2026 13:37:39 +0000https://ramdi.fr/github-stars/clawsync-a-convex-based-multi-agent-ai-platform-with-shared-soul-documents-and-per-agent-model-routing/ClawSync offers a multi-agent AI platform using Convex backend, with shared soul documents for reusable personalities and per-agent model routing across popular LLMs. Explore its architecture and setup.https://ramdi.fr/github-stars/deepdrone-natural-language-ai-control-for-drones-with-real-time-telemetry-and-mavlink-integration/Tue, 05 May 2026 13:37:39 +0000https://ramdi.fr/github-stars/deepdrone-natural-language-ai-control-for-drones-with-real-time-telemetry-and-mavlink-integration/DeepDrone uses LLMs to translate natural language commands into structured drone operations via MAVLink, with real-time telemetry and safety constraints. Python backend, FastAPI, LiteLLM, and JS frontend.https://ramdi.fr/github-stars/inside-company-research-agent-automating-business-intelligence-with-multi-api-ai-agents/Tue, 05 May 2026 13:37:39 +0000https://ramdi.fr/github-stars/inside-company-research-agent-automating-business-intelligence-with-multi-api-ai-agents/Company Research Agent automates detailed business research by orchestrating OpenAI, Google Gemini, Tavily APIs and geolocation data via a Python backend and Node.js frontend. Setup script streamlines install.https://ramdi.fr/github-stars/llm-wiki-orchestrating-multi-agent-llm-research-into-persistent-knowledge-bases/Tue, 05 May 2026 13:37:39 +0000https://ramdi.fr/github-stars/llm-wiki-orchestrating-multi-agent-llm-research-into-persistent-knowledge-bases/llm-wiki is a shell-based orchestration layer that turns LLM agents into a persistent, multi-agent research wiki. Supports up to 10 agents, deep investigations, and durable provenance tracking.https://ramdi.fr/github-stars/nanobrowser-multi-agent-ai-browser-automation-with-dynamic-self-correcting-planning/Tue, 05 May 2026 13:37:39 +0000https://ramdi.fr/github-stars/nanobrowser-multi-agent-ai-browser-automation-with-dynamic-self-correcting-planning/Nanobrowser is a TypeScript Chrome extension implementing a multi-agent AI system for browser automation with a unique self-correcting Planner-Navigator architecture, supporting multiple LLMs and local privacy.https://ramdi.fr/github-stars/pokieticker-layered-ai-driven-stock-market-analysis-with-sentiment-and-xgboost/Tue, 05 May 2026 13:37:39 +0000https://ramdi.fr/github-stars/pokieticker-layered-ai-driven-stock-market-analysis-with-sentiment-and-xgboost/PokieTicker combines rule-based filtering, LLM sentiment analysis, and XGBoost prediction in a full-stack stock analysis app. Runs locally with no API keys.https://ramdi.fr/github-stars/zeron-chat-a-unified-ai-chat-interface-with-resumable-streaming-for-multi-llm-experimentation/Tue, 05 May 2026 13:37:39 +0000https://ramdi.fr/github-stars/zeron-chat-a-unified-ai-chat-interface-with-resumable-streaming-for-multi-llm-experimentation/Zeron Chat is a TypeScript React app that unifies multiple LLM providers in one interface with resumable streaming that survives page refreshes, built on TanStack Start and Zero state management.https://ramdi.fr/github-stars/zinc-a-zig-based-llm-inference-engine-optimized-for-amd-rdna-and-apple-silicon-gpus/Tue, 05 May 2026 13:37:39 +0000https://ramdi.fr/github-stars/zinc-a-zig-based-llm-inference-engine-optimized-for-amd-rdna-and-apple-silicon-gpus/Zinc is a Zig-written LLM inference engine using Vulkan and Metal for AMD RDNA and Apple Silicon GPUs. It supports GGUF quantized models and exposes an OpenAI-compatible API with streaming.https://ramdi.fr/github-stars/ai-trader-ai-powered-config-driven-backtesting-with-natural-language-interaction/Mon, 04 May 2026 10:23:02 +0000https://ramdi.fr/github-stars/ai-trader-ai-powered-config-driven-backtesting-with-natural-language-interaction/ai-trader adds natural language AI interaction to algorithmic trading backtesting via an MCP server and YAML configs. Supports US/TW stocks, crypto, forex with caching.https://ramdi.fr/github-stars/allium-a-behavioral-specification-framework-for-intent-persistence-in-ai-agent-engineering/Mon, 04 May 2026 10:23:02 +0000https://ramdi.fr/github-stars/allium-a-behavioral-specification-framework-for-intent-persistence-in-ai-agent-engineering/Allium addresses intent drift in AI agent sessions by capturing behaviors as formal specs that persist across interactions, exposing contradictions automatically.https://ramdi.fr/github-stars/building-private-ai-workflows-with-the-n8n-self-hosted-ai-starter-kit/Mon, 04 May 2026 10:23:02 +0000https://ramdi.fr/github-stars/building-private-ai-workflows-with-the-n8n-self-hosted-ai-starter-kit/Spin up a private AI agent stack in under 5 minutes with n8n’s self-hosted AI starter kit. Combines local LLMs, automation, and vector search for secure AI workflows.https://ramdi.fr/github-stars/council-of-high-intelligence-orchestrating-structured-multi-agent-ai-deliberations-across-multiple-llms/Mon, 04 May 2026 10:23:02 +0000https://ramdi.fr/github-stars/council-of-high-intelligence-orchestrating-structured-multi-agent-ai-deliberations-across-multiple-llms/Council of High Intelligence is a Shell tool coordinating 18 AI personas across Claude, OpenAI, Gemini, and Ollama, enforcing true disagreement via structured multi-round deliberations.https://ramdi.fr/github-stars/docsgpt-a-flexible-ai-platform-for-private-agents-and-enterprise-document-search/Mon, 04 May 2026 10:23:02 +0000https://ramdi.fr/github-stars/docsgpt-a-flexible-ai-platform-for-private-agents-and-enterprise-document-search/DocsGPT is a Python-based AI platform for building private agents and enterprise search, with multi-LLM support and versatile deployment modes via Docker Compose.https://ramdi.fr/github-stars/exploring-claude-api-integration-patterns-with-anthropics-claude-cookbooks/Mon, 04 May 2026 10:23:02 +0000https://ramdi.fr/github-stars/exploring-claude-api-integration-patterns-with-anthropics-claude-cookbooks/anthropics/claude-cookbooks offers Jupyter Notebook recipes demonstrating practical Claude API usage, including sub-agent orchestration, multimodal vision, and RAG patterns.https://ramdi.fr/github-stars/gitingest-turning-github-repos-into-ai-friendly-text-digests-with-a-clever-url-hack/Mon, 04 May 2026 10:23:02 +0000https://ramdi.fr/github-stars/gitingest-turning-github-repos-into-ai-friendly-text-digests-with-a-clever-url-hack/Gitingest is a Python CLI and API that converts Git repos into LLM-optimized text digests, featuring a unique URL hack for instant GitHub repo ingestion and self-hosted FastAPI server.https://ramdi.fr/github-stars/hands-on-large-language-models-a-practical-visual-journey-through-llm-engineering/Mon, 04 May 2026 10:23:02 +0000https://ramdi.fr/github-stars/hands-on-large-language-models-a-practical-visual-journey-through-llm-engineering/Explore the Hands-On Large Language Models repo, a Jupyter notebook-based practical guide from fundamentals to fine-tuning, designed for hands-on LLM learning on free Colab GPUs.https://ramdi.fr/github-stars/hf-agents-a-shell-cli-extension-for-hardware-aware-local-coding-agents-with-llama-cpp/Mon, 04 May 2026 10:23:02 +0000https://ramdi.fr/github-stars/hf-agents-a-shell-cli-extension-for-hardware-aware-local-coding-agents-with-llama-cpp/hf-agents automates hardware profiling, model selection, and local coding agent deployment using llama.cpp and Pi, all in a shell CLI extension. Efficient and minimal dependencies.https://ramdi.fr/github-stars/how-the-claude-plugins-repo-orchestrates-multi-agent-ai-consultation-with-multiple-llms/Mon, 04 May 2026 10:23:02 +0000https://ramdi.fr/github-stars/how-the-claude-plugins-repo-orchestrates-multi-agent-ai-consultation-with-multiple-llms/claude-plugins is a TypeScript-based plugin marketplace for Claude Code, featuring a multi-agent consultant plugin that runs parallel LLMs like GPT-5, Gemini, Grok, Perplexity, and Claude for AI consultation.https://ramdi.fr/github-stars/how-video-use-turns-ai-agents-into-transcript-driven-video-editors/Mon, 04 May 2026 10:23:02 +0000https://ramdi.fr/github-stars/how-video-use-turns-ai-agents-into-transcript-driven-video-editors/video-use replaces frame-heavy editing with transcript-driven AI agents, using ElevenLabs Scribe and self-evaluation to produce polished edits.https://ramdi.fr/github-stars/inside-nousresearch-s-finetuning-subnet-continuous-incentivized-fine-tuning-for-llms-on-bittensor/Mon, 04 May 2026 10:23:02 +0000https://ramdi.fr/github-stars/inside-nousresearch-s-finetuning-subnet-continuous-incentivized-fine-tuning-for-llms-on-bittensor/NousResearch’s finetuning-subnet enables continuous, incentivized fine-tuning of LLMs using synthetic data from a separate subnet, pioneering cross-subnet communication in Bittensor.https://ramdi.fr/github-stars/isc-bench-exposing-fundamental-ai-safety-failures-from-workflow-level-design/Mon, 04 May 2026 10:23:02 +0000https://ramdi.fr/github-stars/isc-bench-exposing-fundamental-ai-safety-failures-from-workflow-level-design/ISC-Bench reveals a structural AI safety flaw where LLMs produce harmful outputs to complete tasks, bypassing prompt-level defenses. It benchmarks this workflow-level vulnerability across top models.https://ramdi.fr/github-stars/kohakuterrarium-modular-ai-agent-composition-with-algebraic-pipelines/Mon, 04 May 2026 10:23:02 +0000https://ramdi.fr/github-stars/kohakuterrarium-modular-ai-agent-composition-with-algebraic-pipelines/KohakuTerrarium offers a Python framework to build modular AI agents using a unique algebra for composing multi-agent pipelines, with session persistence and multi-runtime support.https://ramdi.fr/github-stars/kvcached-a-plugin-cache-for-sglang-and-vllm-python-environments/Mon, 04 May 2026 10:23:02 +0000https://ramdi.fr/github-stars/kvcached-a-plugin-cache-for-sglang-and-vllm-python-environments/kvcached provides a plugin cache layer for SGLang and vLLM Python LLM environments, easing deployment with PyPI and Docker support. Useful for optimizing LLM workflows.https://ramdi.fr/github-stars/llm-god-orchestrating-multiple-llm-web-uis-in-one-electron-app-with-dom-injection/Mon, 04 May 2026 10:23:02 +0000https://ramdi.fr/github-stars/llm-god-orchestrating-multiple-llm-web-uis-in-one-electron-app-with-dom-injection/LLM-God bundles multiple LLM web interfaces into a single Electron app, using DOM injection to send prompts to all models simultaneously. It offers a clever free-tier workaround with tradeoffs.https://ramdi.fr/github-stars/lucebox-hub-hand-optimized-cuda-kernels-for-efficient-llm-inference-on-rtx-3090-and-beyond/Mon, 04 May 2026 10:23:02 +0000https://ramdi.fr/github-stars/lucebox-hub-hand-optimized-cuda-kernels-for-efficient-llm-inference-on-rtx-3090-and-beyond/Lucebox Hub optimizes LLM inference on consumer GPUs using a megakernel CUDA approach and speculative decoding, achieving high throughput on RTX 3090 and newer Nvidia GPUs.https://ramdi.fr/github-stars/lycheememory-a-lightweight-semantic-long-term-memory-framework-for-llm-agents/Mon, 04 May 2026 10:23:02 +0000https://ramdi.fr/github-stars/lycheememory-a-lightweight-semantic-long-term-memory-framework-for-llm-agents/LycheeMemory offers a lightweight semantic memory system for LLM agents, cutting token use by 71% and costs by 55% compared to native memory, with SQLite + LanceDB backend and REST/MCP APIs.https://ramdi.fr/github-stars/magi-a-structured-multi-llm-debate-system-with-iterative-critique-and-voting/Mon, 04 May 2026 10:23:02 +0000https://ramdi.fr/github-stars/magi-a-structured-multi-llm-debate-system-with-iterative-critique-and-voting/MAGI implements a multi-round debate protocol among three LLMs to match stronger models’ accuracy via iterative critique and voting. It offers fault tolerance, adaptive escalation, and persona presets.https://ramdi.fr/github-stars/mapping-the-llm-agent-landscape-with-the-awesome-llm-agents-curated-catalog/Mon, 04 May 2026 10:23:02 +0000https://ramdi.fr/github-stars/mapping-the-llm-agent-landscape-with-the-awesome-llm-agents-curated-catalog/A curated catalog of 20+ LLM agent frameworks and tools organized by agent type and capabilities. Understand architectural differences and trade-offs in LLM agent design.https://ramdi.fr/github-stars/memary-recursive-knowledge-graph-memory-for-autonomous-ai-agents/Mon, 04 May 2026 10:23:02 +0000https://ramdi.fr/github-stars/memary-recursive-knowledge-graph-memory-for-autonomous-ai-agents/Memary is an open-source memory layer for AI agents using knowledge graphs and recursive retrieval to efficiently store and query agent memories. It supports multi-agent setups and integrates with LlamaIndex and OpenAI.https://ramdi.fr/github-stars/meta-harness-evolving-the-scaffolding-around-large-language-models-for-optimized-task-performance/Mon, 04 May 2026 10:23:02 +0000https://ramdi.fr/github-stars/meta-harness-evolving-the-scaffolding-around-large-language-models-for-optimized-task-performance/Meta-Harness from Stanford IRIS Lab automates the search for optimal harness configurations around LLMs, evolving memory, retrieval, and context systems for better task-specific performance.https://ramdi.fr/github-stars/oasis-a-python-cli-for-ai-driven-code-vulnerability-scanning-with-deterministic-validation/Mon, 04 May 2026 10:23:02 +0000https://ramdi.fr/github-stars/oasis-a-python-cli-for-ai-driven-code-vulnerability-scanning-with-deterministic-validation/OASIS is a Python CLI security auditor using LangGraph-orchestrated LLMs for two-phase scanning and deterministic validation of code vulnerabilities. It balances AI insights with guardrails to reduce false positives.https://ramdi.fr/github-stars/opengame-generating-playable-web-games-from-natural-language-with-a-dual-skill-llm-framework/Mon, 04 May 2026 10:23:02 +0000https://ramdi.fr/github-stars/opengame-generating-playable-web-games-from-natural-language-with-a-dual-skill-llm-framework/OpenGame from CUHK MMLab generates full web games from natural language prompts using a dual-skill LLM architecture that maintains cross-file consistency and integration fixes.https://ramdi.fr/github-stars/openkb-a-persistent-vectorless-wiki-knowledge-base-powered-by-llms-and-pageindex/Mon, 04 May 2026 10:23:02 +0000https://ramdi.fr/github-stars/openkb-a-persistent-vectorless-wiki-knowledge-base-powered-by-llms-and-pageindex/OpenKB compiles documents into a persistent, interlinked wiki using LLMs and PageIndex’s vectorless retrieval, supporting multi-LLM backends and interactive chat with persisted sessions.https://ramdi.fr/github-stars/orion-direct-access-to-apple-neural-engine-for-on-device-llm-training/Mon, 04 May 2026 10:23:02 +0000https://ramdi.fr/github-stars/orion-direct-access-to-apple-neural-engine-for-on-device-llm-training/Orion bypasses CoreML to access Apple’s Neural Engine directly via private frameworks, enabling on-device inference and fine-tuning of small LLMs with 8.5x reduced training overhead.https://ramdi.fr/github-stars/pagelm-orchestrating-multi-provider-llm-workflows-for-interactive-learning/Mon, 04 May 2026 10:23:02 +0000https://ramdi.fr/github-stars/pagelm-orchestrating-multi-provider-llm-workflows-for-interactive-learning/PageLM is an open-source TypeScript platform orchestrating multi-LLM workflows to generate interactive educational content from documents with real-time streaming and multi-backend support.https://ramdi.fr/github-stars/pasteguard-a-local-privacy-proxy-for-masking-sensitive-data-in-llm-requests/Mon, 04 May 2026 10:23:02 +0000https://ramdi.fr/github-stars/pasteguard-a-local-privacy-proxy-for-masking-sensitive-data-in-llm-requests/PasteGuard intercepts API calls to OpenAI and Anthropic, masking over 30 types of sensitive data across 24 languages before reaching AI providers. Simple integration by changing base URL.https://ramdi.fr/github-stars/pdftochat-a-cloud-integrated-pdf-to-chat-system-with-hybrid-vector-search/Mon, 04 May 2026 10:23:02 +0000https://ramdi.fr/github-stars/pdftochat-a-cloud-integrated-pdf-to-chat-system-with-hybrid-vector-search/pdftochat is a TypeScript-based PDF-to-chat app leveraging Chroma Cloud for hybrid vector search and Together.ai for LLMs, integrating multiple cloud services for scalable document Q&A.https://ramdi.fr/github-stars/resume-matcher-a-provider-agnostic-ai-platform-for-tailored-resumes-using-litellm-abstraction/Mon, 04 May 2026 10:23:02 +0000https://ramdi.fr/github-stars/resume-matcher-a-provider-agnostic-ai-platform-for-tailored-resumes-using-litellm-abstraction/Resume Matcher uses LiteLLM to unify six LLM providers for AI-powered resume tailoring, with a FastAPI backend and Next.js frontend. It supports local and cloud deployments with PDF export.https://ramdi.fr/github-stars/smallclaw-a-local-first-ai-agent-framework-with-single-pass-chat-handling/Mon, 04 May 2026 10:23:02 +0000https://ramdi.fr/github-stars/smallclaw-a-local-first-ai-agent-framework-with-single-pass-chat-handling/SmallClaw is a TypeScript AI agent framework that uses a single LLM call for chat and tool invocation, designed for local models with a clean web UI and structured tools.https://ramdi.fr/github-stars/textgen-a-portable-zero-config-local-llm-runner-with-multi-backend-and-multimodal-support/Mon, 04 May 2026 10:23:02 +0000https://ramdi.fr/github-stars/textgen-a-portable-zero-config-local-llm-runner-with-multi-backend-and-multimodal-support/TextGen offers a portable desktop app for local LLMs with zero telemetry and multi-backend support. Drop GGUF models in a folder and run with no complex setup. It features multimodal vision, file attachments, and OpenAI-compatible API.https://ramdi.fr/github-stars/understanding-llm-internals-a-hands-on-guide-to-transformers-and-attention-math/Mon, 04 May 2026 10:23:02 +0000https://ramdi.fr/github-stars/understanding-llm-internals-a-hands-on-guide-to-transformers-and-attention-math/A curated repo breaking down large language model internals with numeric attention math, tokenization, and transformer architecture, targeting engineers who want to understand LLMs under the hood.https://ramdi.fr/github-stars/vllm-mlx-efficient-llm-serving-on-apple-silicon-with-ssd-tiered-kv-cache-and-continuous-batching/Mon, 04 May 2026 10:23:02 +0000https://ramdi.fr/github-stars/vllm-mlx-efficient-llm-serving-on-apple-silicon-with-ssd-tiered-kv-cache-and-continuous-batching/vllm-mlx is a Python inference server for Apple Silicon that supports OpenAI and Anthropic APIs, featuring SSD-tiered KV cache for long-context agents and continuous batching for performance.https://ramdi.fr/github-stars/writehere-dynamic-recursive-planning-for-ai-assisted-long-form-writing-with-real-time-visualization/Mon, 04 May 2026 10:23:02 +0000https://ramdi.fr/github-stars/writehere-dynamic-recursive-planning-for-ai-assisted-long-form-writing-with-real-time-visualization/WriteHERE uses recursive task decomposition to dynamically break down and execute long-form AI writing tasks, with real-time visualization of the agent’s thought process. Supports GPT and Claude.https://ramdi.fr/github-stars/a-practical-taxonomy-for-large-language-model-ensembles-exploring-the-awesome-llm-ensemble-repository/Mon, 04 May 2026 10:23:01 +0000https://ramdi.fr/github-stars/a-practical-taxonomy-for-large-language-model-ensembles-exploring-the-awesome-llm-ensemble-repository/The Awesome-LLM-Ensemble repo catalogs research on combining multiple LLMs with a clear three-phase taxonomy: before, during, and after inference ensemble methods.https://ramdi.fr/github-stars/ai-knowledge-graph-generator-building-structured-graphs-from-unstructured-text-with-llms/Mon, 04 May 2026 10:23:01 +0000https://ramdi.fr/github-stars/ai-knowledge-graph-generator-building-structured-graphs-from-unstructured-text-with-llms/A Python tool that converts unstructured text into interactive knowledge graphs using a three-phase LLM pipeline with SPO triplet extraction, entity standardization, and relationship inference.https://ramdi.fr/github-stars/automating-bank-statement-processing-with-yolov8-ocr-and-llms-for-personal-finance-analysis/Mon, 04 May 2026 10:23:01 +0000https://ramdi.fr/github-stars/automating-bank-statement-processing-with-yolov8-ocr-and-llms-for-personal-finance-analysis/Explore how a hybrid pipeline using YOLOv8 layout detection, OCR, and LLMs automates messy bank statement PDFs for personal finance analysis with RAG and AI agents.https://ramdi.fr/github-stars/boxpwnr-benchmarking-autonomous-llm-agents-on-cybersecurity-challenges-with-iterative-command-execution/Mon, 04 May 2026 10:23:01 +0000https://ramdi.fr/github-stars/boxpwnr-benchmarking-autonomous-llm-agents-on-cybersecurity-challenges-with-iterative-command-execution/BoxPwnr benchmarks LLM-based autonomous agents on cybersecurity challenges using iterative command execution in a Kali Docker container, supporting 20+ LLM models and 13+ platforms.https://ramdi.fr/github-stars/curating-quality-a-curated-list-of-essential-books-for-large-language-model-engineers/Mon, 04 May 2026 10:23:01 +0000https://ramdi.fr/github-stars/curating-quality-a-curated-list-of-essential-books-for-large-language-model-engineers/A curated list of 24 rigorously selected books on LLM engineering, covering foundational theory to production deployment. Highlights a unique 6-step quality filtering process.https://ramdi.fr/github-stars/datagen-a-langgraph-multi-agent-framework-for-automated-data-analysis-workflows/Mon, 04 May 2026 10:23:01 +0000https://ramdi.fr/github-stars/datagen-a-langgraph-multi-agent-framework-for-automated-data-analysis-workflows/DATAGEN orchestrates eight specialized AI agents using LangGraph to automate data analysis workflows with progressive disclosure and multi-LLM provider support.https://ramdi.fr/github-stars/docstrange-a-versatile-python-library-for-llm-optimized-document-parsing-with-dual-mode-processing/Mon, 04 May 2026 10:23:01 +0000https://ramdi.fr/github-stars/docstrange-a-versatile-python-library-for-llm-optimized-document-parsing-with-dual-mode-processing/DocStrange converts PDFs, DOCX, PPTX, XLSX, images, and URLs into LLM-ready Markdown, JSON, HTML, and CSV. It offers free cloud and private local GPU modes for flexible, privacy-compliant document parsing.https://ramdi.fr/github-stars/eclaire-a-local-first-ai-assistant-unifying-your-personal-data-with-local-llms/Mon, 04 May 2026 10:23:01 +0000https://ramdi.fr/github-stars/eclaire-a-local-first-ai-assistant-unifying-your-personal-data-with-local-llms/Eclaire is a self-hosted AI assistant that unifies personal data with local LLM backends via an OpenAI-compatible API, emphasizing privacy and modular design.https://ramdi.fr/github-stars/inside-llm-madness-a-lightweight-gpt-transformer-training-pipeline-with-built-in-visualization/Mon, 04 May 2026 10:23:01 +0000https://ramdi.fr/github-stars/inside-llm-madness-a-lightweight-gpt-transformer-training-pipeline-with-built-in-visualization/llm-madness offers a Python-built GPT-style transformer training pipeline with tokenizer training, memory-mapped datasets, and a unique web UI for per-layer attention inspection and loss visualization.https://ramdi.fr/github-stars/mind-map-wizard-ai-powered-mind-maps-with-a-custom-svg-rendering-engine/Mon, 04 May 2026 10:23:01 +0000https://ramdi.fr/github-stars/mind-map-wizard-ai-powered-mind-maps-with-a-custom-svg-rendering-engine/Mind-Map-Wizard generates interactive mind maps from AI-generated markdown outlines using a custom SVG engine and keeps all data local for privacy.https://ramdi.fr/github-stars/openclaude-a-multi-model-terminal-first-coding-agent-cli-with-practical-agent-routing/Mon, 04 May 2026 10:23:01 +0000https://ramdi.fr/github-stars/openclaude-a-multi-model-terminal-first-coding-agent-cli-with-practical-agent-routing/OpenClaude is a TypeScript CLI coding agent that routes tasks across different LLMs by type, optimizing cost and performance with multi-provider support and a unified terminal interface.https://ramdi.fr/github-stars/openmythos-exploring-recurrent-depth-transformers-with-input-injection-for-sustained-reasoning/Mon, 04 May 2026 10:23:01 +0000https://ramdi.fr/github-stars/openmythos-exploring-recurrent-depth-transformers-with-input-injection-for-sustained-reasoning/OpenMythos implements a recurrent-depth transformer that recycles layers via looped blocks, using input injection to prevent signal drift. It scales from 1B to 1T parameters with up to 1M token context.https://ramdi.fr/github-stars/pat3d-orchestrating-text-to-3d-simulation-ready-scenes-through-a-multi-stage-ai-and-physics-pipeline/Mon, 04 May 2026 10:23:01 +0000https://ramdi.fr/github-stars/pat3d-orchestrating-text-to-3d-simulation-ready-scenes-through-a-multi-stage-ai-and-physics-pipeline/PAT3D composes a 9-stage pipeline combining LLMs, vision models, 3D asset generators, and physics simulation to produce physically plausible, simulation-ready 3D scenes from text prompts.https://ramdi.fr/github-stars/quantaalpha-llm-driven-trajectory-based-self-evolution-for-quantitative-alpha-factor-discovery/Mon, 04 May 2026 10:23:01 +0000https://ramdi.fr/github-stars/quantaalpha-llm-driven-trajectory-based-self-evolution-for-quantitative-alpha-factor-discovery/QuantaAlpha uses large language models with evolutionary strategies to automate quantitative alpha factor discovery, achieving strong backtest metrics on major indices.https://ramdi.fr/github-stars/ray-finance-a-local-first-privacy-focused-cli-financial-advisor-with-encrypted-context-and-llm-powered-advice/Mon, 04 May 2026 10:23:01 +0000https://ramdi.fr/github-stars/ray-finance-a-local-first-privacy-focused-cli-financial-advisor-with-encrypted-context-and-llm-powered-advice/ray-finance is a TypeScript CLI tool that syncs bank data locally with AES-256 encryption, redacts PII before AI calls, and maintains persistent financial context for personalized LLM advice.https://ramdi.fr/github-stars/restai-a-multi-project-aiaas-platform-with-agentic-browser-automation-and-visual-ai-pipelines/Mon, 04 May 2026 10:23:01 +0000https://ramdi.fr/github-stars/restai-a-multi-project-aiaas-platform-with-agentic-browser-automation-and-visual-ai-pipelines/RESTai exposes multi-project AI capabilities via a unified REST API, featuring an agentic browser with Dockerized Playwright, knowledge graph RAG, and a visual Blockly pipeline builder.https://ramdi.fr/github-stars/rsclaw-a-rust-native-ai-agent-engine-with-persistent-three-layer-memory-and-multi-agent-delegation/Mon, 04 May 2026 10:23:01 +0000https://ramdi.fr/github-stars/rsclaw-a-rust-native-ai-agent-engine-with-persistent-three-layer-memory-and-multi-agent-delegation/RsClaw is a Rust-based AI agent engine featuring persistent three-layer memory across sessions, multi-agent delegation, and low resource usage in a single 15MB binary.https://ramdi.fr/github-stars/stash-a-shared-agent-memory-with-no-server-side-llm-calls/Mon, 04 May 2026 10:23:01 +0000https://ramdi.fr/github-stars/stash-a-shared-agent-memory-with-no-server-side-llm-calls/Stash captures coding agent session transcripts for teams and builds a shared knowledge base without server-side LLM calls, preserving privacy and cutting costs.https://ramdi.fr/github-stars/swarmvault-a-local-first-knowledge-compiler-with-contradiction-detection-and-hybrid-search/Mon, 04 May 2026 10:23:01 +0000https://ramdi.fr/github-stars/swarmvault-a-local-first-knowledge-compiler-with-contradiction-detection-and-hybrid-search/SwarmVault compiles raw sources into a persistent Markdown wiki with typed knowledge graph, hybrid search, and contradiction detection. It supports 16+ agents and offline use.https://ramdi.fr/github-stars/unslop-empirically-detecting-and-avoiding-repetitive-llm-output-patterns/Mon, 04 May 2026 10:11:02 +0000https://ramdi.fr/github-stars/unslop-empirically-detecting-and-avoiding-repetitive-llm-output-patterns/unslop is a Python CLI tool that detects repetitive defaults in LLM outputs by empirical analysis, generating reusable anti-pattern profiles to improve prompt engineering.https://ramdi.fr/github-stars/ai-penetration-testing-knowledge-base-structured-resources-for-llm-security-research/Mon, 04 May 2026 10:09:00 +0000https://ramdi.fr/github-stars/ai-penetration-testing-knowledge-base-structured-resources-for-llm-security-research/A curated repository for AI/LLM penetration testing covering prompt injection, adversarial ML, and LLM red teaming with the OWASP LLM Top 10 framework.https://ramdi.fr/github-stars/inside-llm-wiki-a-desktop-app-for-building-persistent-llm-powered-personal-wikis/Mon, 04 May 2026 10:05:49 +0000https://ramdi.fr/github-stars/inside-llm-wiki-a-desktop-app-for-building-persistent-llm-powered-personal-wikis/llm_wiki uses a two-step chain-of-thought pipeline to build a self-maintaining knowledge base. It combines Tauri, knowledge graphs, and Louvain clustering for a unique personal wiki experience.https://ramdi.fr/github-stars/building-a-production-ready-second-brain-with-agentic-rag-and-llmops/Sun, 03 May 2026 08:12:11 +0000https://ramdi.fr/github-stars/building-a-production-ready-second-brain-with-agentic-rag-and-llmops/Explore an open-source course that teaches building a production-grade AI assistant using advanced retrieval-augmented generation, agent orchestration, fine-tuning, and LLMOps practices.https://ramdi.fr/github-stars/navigating-free-tier-llm-apis-with-the-awesome-free-llm-apis-catalog/Sun, 03 May 2026 08:12:11 +0000https://ramdi.fr/github-stars/navigating-free-tier-llm-apis-with-the-awesome-free-llm-apis-catalog/A curated catalog of free-tier LLM APIs compatible with OpenAI SDK, detailing rate limits, model specs, and providers to build zero-cost AI applications.https://ramdi.fr/github-stars/a-mem-dynamic-semantic-memory-management-for-llm-agents-inspired-by-zettelkasten/Sun, 03 May 2026 00:54:10 +0000https://ramdi.fr/github-stars/a-mem-dynamic-semantic-memory-management-for-llm-agents-inspired-by-zettelkasten/A-MEM is a Python agentic memory system that dynamically organizes LLM agent memories using semantic embeddings and automatic linking, inspired by Zettelkasten.https://ramdi.fr/github-stars/a-hands-on-course-for-mastering-large-language-models-fine-tuning-quantization-and-tooling/Sat, 02 May 2026 20:07:04 +0000https://ramdi.fr/github-stars/a-hands-on-course-for-mastering-large-language-models-fine-tuning-quantization-and-tooling/Explore a comprehensive LLM course with practical notebooks on fine-tuning (QLoRA, DPO), quantization (GPTQ), and tools like AutoEval and LazyMergekit. Ideal for aspiring LLM engineers.https://ramdi.fr/github-stars/hermes-agent-a-self-improving-ai-agent-with-closed-learning-loops-and-multi-platform-integration/Sat, 02 May 2026 20:07:04 +0000https://ramdi.fr/github-stars/hermes-agent-a-self-improving-ai-agent-with-closed-learning-loops-and-multi-platform-integration/Hermes Agent is a Python AI agent featuring closed learning loops, autonomous skill creation, multi-model support, and seamless Telegram/Discord integration for persistent, adaptable AI workflows.https://ramdi.fr/github-stars/llamafactory-modular-extensible-fine-tuning-framework-for-large-language-models/Sat, 02 May 2026 20:07:04 +0000https://ramdi.fr/github-stars/llamafactory-modular-extensible-fine-tuning-framework-for-large-language-models/LlamaFactory offers a modular Python framework for fine-tuning 100+ LLMs with diverse algorithms and optimizations, including LoRA, QLoRA, and reinforcement learning.https://ramdi.fr/github-stars/localai-running-diverse-ai-models-locally-with-multi-backend-support-and-agent-capabilities/Sat, 02 May 2026 20:07:04 +0000https://ramdi.fr/github-stars/localai-running-diverse-ai-models-locally-with-multi-backend-support-and-agent-capabilities/LocalAI enables running 36+ AI models locally without GPU, supporting multi-user API access and built-in AI agents with OpenAI-compatible APIs. Here’s how it works and why it matters.https://ramdi.fr/github-stars/mem0-optimizing-ai-agent-memory-with-a-new-single-pass-additive-algorithm/Sat, 02 May 2026 20:07:04 +0000https://ramdi.fr/github-stars/mem0-optimizing-ai-agent-memory-with-a-new-single-pass-additive-algorithm/mem0 enhances AI agent memory with a new single-pass ADD-only extraction algorithm and multi-signal retrieval, boosting benchmarks significantly while simplifying memory management.https://ramdi.fr/github-stars/metagpt-orchestrating-multi-agent-ai-teams-to-automate-software-development/Sat, 02 May 2026 20:07:04 +0000https://ramdi.fr/github-stars/metagpt-orchestrating-multi-agent-ai-teams-to-automate-software-development/MetaGPT uses a multi-agent system with defined GPT roles following SOPs to automate software development from one-line prompts. It simulates a software company with role-based AI collaboration.https://ramdi.fr/github-stars/ollama-a-unified-cli-and-api-platform-for-local-large-language-models/Sat, 02 May 2026 20:07:04 +0000https://ramdi.fr/github-stars/ollama-a-unified-cli-and-api-platform-for-local-large-language-models/Ollama simplifies running and managing open-source large language models locally with a unified CLI and REST API, supporting broad integrations and multi-OS support.https://ramdi.fr/github-stars/vllm-efficient-large-language-model-serving-with-paged-attention-and-continuous-batching/Sat, 02 May 2026 20:07:04 +0000https://ramdi.fr/github-stars/vllm-efficient-large-language-model-serving-with-paged-attention-and-continuous-batching/vLLM is a Python library for high-throughput LLM inference using paged attention and continuous batching. It supports quantization, distributed inference, and an OpenAI-compatible API.https://ramdi.fr/github-stars/tradingagents-a-multi-agent-llm-framework-simulating-real-world-trading-firm-dynamics/Sat, 02 May 2026 07:48:10 +0000https://ramdi.fr/github-stars/tradingagents-a-multi-agent-llm-framework-simulating-real-world-trading-firm-dynamics/TradingAgents uses specialized LLM agents in a structured bull/bear debate to mimic real trading firms. Supports 10+ LLMs, persistent memory, and CLI/Docker usage.https://ramdi.fr/github-stars/qwen-code-a-multi-provider-terminal-ai-coding-agent-with-unified-config-abstraction/Tue, 28 Apr 2026 18:38:54 +0000https://ramdi.fr/github-stars/qwen-code-a-multi-provider-terminal-ai-coding-agent-with-unified-config-abstraction/Qwen Code is a TypeScript terminal AI coding agent that abstracts multiple LLM providers behind a unified config, enabling flexible AI workflows with Skills and SubAgents.https://ramdi.fr/post/ai-llm/local-llm-tokens-per-second-benchmark/Tue, 28 Apr 2026 00:00:00 +0000https://ramdi.fr/post/ai-llm/local-llm-tokens-per-second-benchmark/Part 2 of 4: a benchmark journal across nixpkgs llama.cpp, upstream master, and ik_llama.cpp on Qwen3.6-27B. Six hours, four backends, all converging at 66 tok/s — and the physical reason why.https://ramdi.fr/post/ai-llm/local-llm-speculative-decoding-hybrid-ssm/Tue, 28 Apr 2026 00:00:00 +0000https://ramdi.fr/post/ai-llm/local-llm-speculative-decoding-hybrid-ssm/Part 3 of 4: a deep-dive into why speculative decoding silently breaks (or runs anti-economically) on hybrid attention+SSM architectures like Qwen3.6, Mamba-2, and RWKV — and what would need to change upstream to fix it.https://ramdi.fr/post/ai-llm/local-llm-nixos-llama-server-module/Tue, 28 Apr 2026 00:00:00 +0000https://ramdi.fr/post/ai-llm/local-llm-nixos-llama-server-module/Part 4 of 4: the actual NixOS module, llama-pull helper, claude-code-router wiring, and one-line workflow for switching models. Five Nix files for a complete, isolated, rollback-able local LLM service.https://ramdi.fr/post/ai-llm/local-llm-rtx5090-why-nixos/Tue, 28 Apr 2026 00:00:00 +0000https://ramdi.fr/post/ai-llm/local-llm-rtx5090-why-nixos/Part 1 of 4: motivation, hardware, and stack choices for serving Qwen3.6-27B locally on a 32 GB consumer GPU with NixOS, before any benchmarks or trade-offs kick in.https://ramdi.fr/github-stars/forge-a-rust-based-multi-agent-ai-coding-assistant-integrated-into-your-terminal-workflow/Sun, 26 Apr 2026 23:47:28 +0000https://ramdi.fr/github-stars/forge-a-rust-based-multi-agent-ai-coding-assistant-integrated-into-your-terminal-workflow/Forge is a Rust-based AI coding agent with multi-agent architecture and a unique ZSH plugin that intercepts shell commands for seamless terminal integration. It supports 300+ LLM providers.https://ramdi.fr/github-stars/autogen-exploring-multi-agent-ai-orchestration-with-python-in-maintenance-mode/Sun, 26 Apr 2026 17:51:11 +0000https://ramdi.fr/github-stars/autogen-exploring-multi-agent-ai-orchestration-with-python-in-maintenance-mode/AutoGen is a Python framework for building multi-agent AI applications with LLM integration, now in maintenance mode with Microsoft Agent Framework as its successor. Learn its architecture, strengths, and how to get started.https://ramdi.fr/github-stars/context7-injecting-real-time-version-specific-docs-into-llm-workflows/Sun, 26 Apr 2026 17:51:11 +0000https://ramdi.fr/github-stars/context7-injecting-real-time-version-specific-docs-into-llm-workflows/Context7 tackles LLM hallucinations by injecting up-to-date, version-specific library docs directly into AI coding agents’ context via CLI or MCP server integration.https://ramdi.fr/github-stars/deerflow-2-0-orchestrating-multi-agent-ai-workflows-with-flexible-llm-integration/Sun, 26 Apr 2026 17:51:11 +0000https://ramdi.fr/github-stars/deerflow-2-0-orchestrating-multi-agent-ai-workflows-with-flexible-llm-integration/DeerFlow 2.0 is a Python framework for orchestrating AI sub-agents and memory with support for multiple LLMs and execution sandboxes. It uses a modular config and setup wizard for flexible deployment.https://ramdi.fr/github-stars/inside-ai-engineering-hub-a-hands-on-collection-of-production-ready-ai-projects/Sun, 26 Apr 2026 17:51:11 +0000https://ramdi.fr/github-stars/inside-ai-engineering-hub-a-hands-on-collection-of-production-ready-ai-projects/AI Engineering Hub offers 90+ production-ready AI projects spanning LLMs, RAG, AI agents, and MCP, organized by difficulty and real-world use cases.https://ramdi.fr/github-stars/inside-cowagent-an-extensible-autonomous-ai-assistant-with-multi-modal-and-multi-model-architecture/Sun, 26 Apr 2026 17:51:11 +0000https://ramdi.fr/github-stars/inside-cowagent-an-extensible-autonomous-ai-assistant-with-multi-modal-and-multi-model-architecture/CowAgent is an extensible AI assistant framework with autonomous task planning, long-term memory, and multi-modal support. It integrates multiple LLMs and platforms for flexible AI workflows.https://ramdi.fr/github-stars/kong-gateway-a-universal-api-gateway-with-advanced-ai-traffic-routing-and-governance/Sun, 26 Apr 2026 17:51:11 +0000https://ramdi.fr/github-stars/kong-gateway-a-universal-api-gateway-with-advanced-ai-traffic-routing-and-governance/Kong Gateway extends traditional API management with universal LLM API routing, semantic security, and AI-specific features, enabling multi-vendor AI traffic governance in cloud-native environments.https://ramdi.fr/github-stars/llm-driven-browser-automation-with-browser-use-a-hands-on-look/Sun, 26 Apr 2026 17:51:11 +0000https://ramdi.fr/github-stars/llm-driven-browser-automation-with-browser-use-a-hands-on-look/Browser-Use is a Python library enabling LLM-powered AI agents to automate browsers efficiently. It features a custom ChatBrowserUse model and supports cloud and local agents.https://ramdi.fr/github-stars/openhands-modular-architecture-for-flexible-ai-agent-development/Sun, 26 Apr 2026 17:51:11 +0000https://ramdi.fr/github-stars/openhands-modular-architecture-for-flexible-ai-agent-development/OpenHands offers a modular Python platform to build and deploy AI agents with SDK, CLI, GUI, and cloud options. It supports multiple LLMs and self-hosting for enterprises.https://ramdi.fr/github-stars/pathway-llm-app-unified-pipelines-for-scalable-retrieval-augmented-generation-and-ai-search/Sun, 26 Apr 2026 09:31:26 +0000https://ramdi.fr/github-stars/pathway-llm-app-unified-pipelines-for-scalable-retrieval-augmented-generation-and-ai-search/Pathway LLM App provides integrated pipelines for scalable RAG and AI search, combining vector and full-text indexing with real-time sync for Gen AI apps at scale.https://ramdi.fr/github-stars/awesome-llm-apps-a-practical-collection-of-runnable-ai-agent-and-rag-templates/Fri, 24 Apr 2026 18:26:13 +0000https://ramdi.fr/github-stars/awesome-llm-apps-a-practical-collection-of-runnable-ai-agent-and-rag-templates/Awesome LLM Apps offers 100+ runnable AI agent and RAG templates for quick LLM app development. It supports multiple providers and advanced multi-agent patterns with minimal setup.https://ramdi.fr/github-stars/inside-daily-stock-analysis-a-multi-llm-automated-stock-analysis-system/Fri, 24 Apr 2026 18:26:13 +0000https://ramdi.fr/github-stars/inside-daily-stock-analysis-a-multi-llm-automated-stock-analysis-system/daily_stock_analysis combines multi-LLM integration with multi-source financial data to automate stock market decisions across global markets. It features flexible AI provider fallback and multi-channel alerts.https://ramdi.fr/github-stars/browser-harness-a-self-healing-llm-agent-for-browser-automation-via-chrome-devtools/Fri, 24 Apr 2026 07:26:29 +0000https://ramdi.fr/github-stars/browser-harness-a-self-healing-llm-agent-for-browser-automation-via-chrome-devtools/Browser Harness enables LLMs to automate browsers by dynamically generating helper functions using the Chrome DevTools Protocol, with minimal Python code and free remote browsers.https://ramdi.fr/post/ai-llm/building-ai-agents-with-claude/Sat, 11 Apr 2026 00:00:00 +0000https://ramdi.fr/post/ai-llm/building-ai-agents-with-claude/How to leverage Claude Code to build autonomous AI agents that publish content, review code, and manage workflows.