Tag
Explore by tags
Open-source TTS that clones a voice from 3-10s of audio and synthesizes cross-lingual speech in 9 languages and 18+ Chinese dialects. Supports streaming at ~150ms latency with instruction control over emotion, speed, and accent.
Converts PDFs, images, and Office documents into Markdown or JSON for retrieval, extraction, and agent workflows, with OCR, layout analysis, formula handling, and multiple runtime modes.
Local-first runtime for autonomous AI agents that run on-device and stay model-agnostic across OpenAI, Anthropic, Gemini, Grok, and local models. A plugin system adds chat platforms (Discord, Telegram, X), voice, browser automation, RAG, and wallets.
Edits a codebase from natural-language prompts in the terminal, coordinating specialized sub-agents — file picker, planner, editor, reviewer — instead of one model. Beats Claude Code 61% vs 53% on its own evals; agents scriptable in TypeScript.
Packs a Git repository into a single AI-friendly file for easy ingestion by LLMs. Offers per-file and total token counts, optional Tree-sitter compression, secret scanning, and multiple interfaces (CLI, web, browser extension, Docker, MCP) for AI-driven code review and analysis.
Routes each user query to the most suitable agent via a classifier that weighs agent profiles and conversation history, keeping context shared across handoffs. Python and TypeScript, with a SupervisorAgent that runs sub-agents in parallel.
Drives UI automation from screenshots alone: describe steps in natural language and a vision model acts on what it sees, no DOM selectors. One API spans web, Android, iOS, HarmonyOS and desktop; plugs into Playwright/Vitest or runs autonomously.
Open-source platform for autonomous coding agents that work like developers: editing files, running shell commands, browsing the web, and calling APIs in an isolated sandbox. Model-agnostic, with GitHub, Slack, and CI/CD integration.
Runs huge mixture-of-experts LLMs like DeepSeek-R1/V3 on a single 24GB GPU plus CPU DRAM by keeping attention on the GPU and offloading expert weights to CPU. Reports 3-28x speedups via Intel AMX/AVX512 kernels and fits 139K context in 24GB VRAM.
Connects any LLM to internal knowledge sources and lets teams chat with cited, RAG-style answers. Notable for broad connectors (Drive, Notion, GitHub, YouTube), universal LLM/embedding support, and self-hostable Docker deployment — aimed at teams that need private, searchable LLM-backed knowledge.
Turns a UI screenshot into structured elements so a vision LLM can act without HTML or accessibility trees. A fine-tuned detector finds interactable icons; a caption model describes their function, lifting GPT-4V grounding on ScreenSpot and Mind2Web.
Official inference framework for 1-bit and ternary (1.58-bit) LLMs such as BitNet b1.58, with optimized CPU kernels. Delivers 1.37x-6.17x speedups and 55-82% lower energy on x86 and ARM, and runs a 100B model on a single CPU at 5-7 tokens/sec.