Category
Explore by categories
Fused CUDA kernels that compute exact attention without ever writing the full N×N score matrix to GPU memory, cutting memory from quadratic to linear and speeding up training and inference on A100/H100. Ships FlashAttention-2/3 plus KV-cache decode paths.
Collects metrics, distributed traces, and continuous profiles via eBPF with zero code instrumentation, covering apps in any language plus gateways, service meshes, databases, and queues. Profiling adds under 1% overhead.
Smart model router for personal AI agents that sends each request to the cheapest model capable of handling it — cutting API costs by up to ~70%. Uses a fast 23-dimension scorer, automatic fallbacks, per-tier controls, and supports local Docker self-hosting or a cloud app; ideal for cost-sensitive personal agents.
Build, run, and monitor LLM agents across one stack: an open framework for chaining models and tools, LangGraph for stateful agent orchestration, and LangSmith for tracing, evaluation, and deployment in production.
Build full‑stack web apps entirely in Python — write frontend components and backend state as Python classes with a reactive model. Provides fast refresh, deployment tooling, and AI-focused integrations such as an AI Builder and an Agent Toolkit for connecting LLMs and image models.
Connects LLMs to private and domain-specific data with ingestion, indexing, and retrieval primitives for RAG and agentic apps. Centers on document parsing via LlamaParse for 90+ file formats, schema-based extraction, and composable queries.
Unified Python framework where the same code runs on batch and streaming data, backed by a Rust engine on Differential Dataflow for incremental computation. Aimed at ETL, analytics, and live RAG pipelines over Kafka and 300+ connectors.
Connects one LLM agent to 15+ chat platforms — QQ, WeChat Work, Feishu, Telegram, Discord, Slack — from a single self-hosted backend. Routes to OpenAI, Anthropic, Gemini, DeepSeek or Ollama, and adds a WebUI, MCP tools, and a 1000+ plugin marketplace.
Lets you write compositional Python programs that compile into self‑improving LLM pipelines — replacing brittle prompt engineering with a declarative, programmatic approach and built‑in algorithms to optimize prompts and weights for RAG, multi‑stage pipelines, and agent loops.
Routes one API call across hundreds of LLMs from dozens of providers, with credits, fallbacks, pricing comparison, and data-policy controls for teams that need model choice without wiring every provider separately.
Aggregates alerts from dozens of monitoring tools into a single pane of glass, then deduplicates, correlates, and enriches them. Automates incident response with declarative YAML workflows — like GitHub Actions for your monitoring stack.
Provides a minimal, Zig-written headless browser tailored for AI agents and automation — runs JavaScript, supports key Web APIs, exposes the Chrome DevTools Protocol for Puppeteer/Playwright, and targets low memory usage and fast startup for large-scale scraping and agent workflows.