AI Image2024
Segments each PDF page into 11 labeled regions — titles, tables, formulas, figures, footnotes and more — and recovers reading order. Offers two engines: an accurate VGT visual model (~0.96 F1) or a faster CPU-only LightGBM ensemble.
AI Train2024
Streamlines the full lifecycle of foundation models — data prep, fine-tuning (SFT/LoRA/QLoRA/GRPO), evaluation, and deployment — with ready-to-run recipes, multi-engine inference support, and cloud/CLI workflows for both laptop experiments and large-scale runs.
AI API2024
Ingests documents, images, audio, video and web pages and converts them into structured, LLM-friendly markdown and parsed data. Runs locally (fits on a T4 GPU), supports ~20 file types, offers OCR, transcription, table extraction and a Gradio UI; deployable via Docker/Skypilot. Licensed under GPL-3.0; some model weights carry cc-by-nc-sa restrictions for commercial use.
AI Infra2024
Disaggregated LLM serving architecture that splits prefill and decode into separate clusters and pools spare CPU, DRAM, and SSD into a distributed KVCache. Powers Kimi in production, handling 75% more requests under the same SLOs.
AI Agent2024
Crawls 30+ social platforms (Weibo, Xiaohongshu, Douyin), parses their video and image content, then has five specialized agents debate in a moderated forum to synthesize public-opinion reports. Can fuse public sentiment with a private business database.
AI Agent2024
Multi-tenant agent harness that makes enterprise knowledge retrievable, graph-reasonable, and deliverable by LLM-powered agents. Integrates RAG + a Milvus-based knowledge graph, LangGraph orchestration, and document parsing for citation-backed answers and graph reasoning; deployable via Docker (requires a compatible LLM API).
AI Client2024
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.
AI Audio2024
Chains four swappable open modules — voice activity detection, speech-to-text, an LLM, and text-to-speech — into a local voice agent that needs no proprietary APIs. Runs on CUDA, Apple Silicon, or Docker, with an OpenAI-compatible realtime WebSocket mode.
AI Deploy2024
Runs and optimizes ML and generative-AI models on-device across mobile, desktop, web, and IoT. Successor to TensorFlow Lite, it adds automated GPU/NPU accelerator selection and zero-copy buffer interop to cut latency without cloud round-trips.
AI Others2024
Python web scraping framework that automatically relocates elements when a site's HTML changes, so selectors survive redesigns. Bundles Cloudflare Turnstile bypass, TLS fingerprint impersonation, and a Scrapy-like async spider for full crawls.
Embodied AI2024
Agentive operating system for physical robots that lets developers compose agent-native modules in Python to connect perception, spatial memory, and control across humanoids, quadrupeds, drones, and simulators.
AI Client2024
Self-hostable alternative to Google NotebookLM: organize PDFs, videos, audio, web pages, and Office docs, then chat over them, take AI-assisted notes, and search via full-text and vector. Routes to 18+ model providers and generates 1-4 speaker podcasts.