Tag
Explore by tags
Terminal coding agent forked from Google's Gemini CLI, retuned for Qwen3-Coder with a custom parser and tool protocol. Runs against OpenAI, Anthropic, Gemini, Qwen or local models, and adds subagents, agent teams, auto-memory and MCP.
Source code for the GitHub Copilot Chat extension in VS Code: inline chat, an agent mode that plans and edits files autonomously, next-edit suggestions, and MCP integration. Open-sourced so developers can study how Copilot connects to an editor.
Teaches agent harness engineering — the permissions, memory, persistence, and coordination layer that lets an LLM act — across 20 progressive lessons, each adding one mechanism with standalone runnable code. Chinese-first, plus English and Japanese.
Stores a pruned proximity graph instead of all embeddings, recomputing vectors on demand at query time. A 60M-doc index takes 6GB, not 201GB (97% less), at comparable recall. Powers private local RAG over files, mail, chat, and browser history.
A template and workflow for feeding AI coding assistants structured context — project rules, code examples, and validation gates — instead of one-off prompts. Centers on Product Requirements Prompts (PRPs) that an agent generates, then executes.
Official, runnable examples for Amazon Bedrock AgentCore, AWS's framework- and model-agnostic platform for deploying AI agents. Spans Runtime, Memory, Gateway, Identity, and Observability through notebooks, code, and infrastructure templates.
Model-compression toolkit for large LLMs/VLMs that integrates quantization (FP8/INT4/etc.), speculative decoding, token pruning and deployment hooks—designed for end-to-end performance on single/multi-GPU inference workflows and research-to-prod model optimization.
Extracts structured data from unstructured text with LLMs, mapping every extraction to its exact character span in the source for visual review. Uses few-shot examples, schema enforcement, and multi-pass chunking to handle long documents.
Runs a six-month live experiment where ChatGPT manages a real-money micro-cap portfolio from $100, trading under strict rules with automated stop-losses. Each trade's rationale is logged; returns are benchmarked against the S&P 500 and Russell 2000.
Spec-driven agentic dev platform that turns a prompt into requirements, a design doc, and sequenced tasks before any code is written, then implements from the spec. Runs across IDE, CLI, web, and mobile; validates output with property-based tests.
An agentic framework that analyzes, plans, and executes multi-step video understanding and editing workflows using multimodal LLM-driven agents—features intent decomposition, graph-based workflow orchestration, and automated shot planning for long-form video tasks.
Unifies agentic tasks, reasoning, and coding in a single MoE model with 355B total / 32B active parameters and a switchable thinking mode. A lighter 106B-param Air variant trades scale for efficiency; both ship MIT-licensed.