Notebooks and sample apps demonstrating generative-AI workflows on Google Cloud's Vertex AI and Gemini — covering RAG grounding, multimodal demos, function calling, and agent-building examples, with deployment-ready templates for evaluation and production.
Evaluates and tests LLM apps — RAG pipelines, agents, and workflows — using objective metrics that mix LLM-as-judge scoring with deterministic measures. Auto-generates synthetic test datasets and integrates with LangChain and tracing tools.
Provides end-to-end observability, evaluation, and optimization for LLM-based applications by tracing model calls, running automated evaluations, and surfacing production metrics. Ships SDKs, broad framework integrations, LLM-as-a-judge metrics, and dashboards to support development, CI, and production monitoring.
Contains short, small-vocabulary stories synthetically generated by GPT-3.5 and GPT-4 for training and evaluating compact language models. Includes multiple splits, a GPT-4-only V2 subset, and archive files with prompts and metadata for reproducible experiments.
Edits code across an existing repo from the terminal: you describe a change in plain English, it maps the whole codebase, applies edits to the right files, and auto-commits each change as a reviewable git commit. Works with most LLMs.
Synchronized desktop chat browser that opens multiple LLM webapps (ChatGPT, Claude, Bard, Bing, Llama2) and submits the same prompt to each for fast cross-provider comparison. Offers keyboard shortcuts, local-model hooks, and prompt-improvement utilities.
Tracks, evaluates, and debugs LLM applications with traces, prompt management, datasets, playgrounds, and observability that can run in cloud or self-hosted setups.
Self-hostable chat client that unifies many LLM providers (OpenAI, Claude, Gemini, Ollama, DeepSeek) behind one UI. Adds file-upload knowledge-base RAG, vision/TTS, an MCP plugin system, and an agent marketplace, with one-click Vercel or Docker deployment.
Unifies access to OpenAI, Anthropic, Google and other LLM providers behind one TypeScript API — swap models by changing a string. Adds streaming UI hooks for React, Next.js, Svelte and Vue, plus a tool-calling loop for agentic workflows.
Browser-first, local-first note app that stores your notes as plain Markdown files and offers a chat-like input flow. LLM-friendly file structure, offline-capable PWA, and Telegram bot integration — data stays on your device unless you enable syncing.
Runs retrieval-augmented Q&A over your own documents on local hardware, so files never leave your machine. Blends semantic, keyword, and late-chunking retrieval, with a router that picks RAG or a direct LLM answer per query and verifies it.
Fine-tunes 100+ LLMs and VLMs from one config file or a no-code web UI, unifying LoRA, QLoRA, full tuning, DPO, PPO, KTO and ORPO behind a single interface. Bundles GaLore, Unsloth, FlashAttention-2 and 2-8bit quantization to fit a single 24GB GPU.