bolt.diy
bolt.diy is an open-source platform that enables developers to easily create, run, edit, and deploy full-stack web applications with a variety of large language models (LLMs). It supports a wide range of models, including OpenAI, Anthropic, Ollama, OpenRouter, Gemini, LMStudio, Mistral, xAI, HuggingFace, DeepSeek, and Groq. The platform offers seamless integration through the Vercel AI SDK, allowing users to customize and extend their applications with the LLMs of their choice. With its intuitive interface, bolt.diy is designed to simplify AI development workflows, making it a great tool for both experimentation and production-ready applications.
Learn more
Martian
By using the best-performing model for each request, we can achieve higher performance than any single model. Martian outperforms GPT-4 across OpenAI's evals (open/evals). We turn opaque black boxes into interpretable representations. Our router is the first tool built on top of our model mapping method. We are developing many other applications of model mapping including turning transformers from indecipherable matrices into human-readable programs. If a company experiences an outage or high latency period, automatically reroute to other providers so your customers never experience any issues. Determine how much you could save by using the Martian Model Router with our interactive cost calculator. Input your number of users, tokens per session, and sessions per month, and specify your cost/quality tradeoff.
Learn more
Arcade
Arcade.dev is an AI tool-calling platform that enables AI agents to securely perform real-world actions, like sending emails, messaging, updating systems, or triggering workflows, through authenticated, user-authorized integrations. By acting as an authenticated proxy based on the OpenAI API spec, Arcade.dev lets models invoke external services (such as Gmail, Slack, GitHub, Salesforce, Notion, and more) via pre-built connectors or custom tool SDKs, managing authentication, token handling, and security seamlessly. Developers work with a unified client interface (arcadepy for Python or arcadejs for JavaScript), facilitating tool execution and authorization without burdening application logic with credentials or API specifics. It supports secure deployments in the cloud, private VPCs, or on premises, and includes a control plane for managing tools, users, permissions, and observability.
Learn more
Kerlig
Kerlig is an AI-powered writing assistant for Mac that helps users save time and improve their communication at work by integrating with all apps. It supports multi-language features and allows you to proofread, summarize, translate, and extract key points from PDFs, documents, and web pages.
Custom Actions and presets make it adaptable to your workflow.
Custom Actions allow for editing prompts to make the AI model perform exactly what you need. Invoke them with a single click or a keyboard shortcut.
Presets are personalized settings for AI models that give them a specific personality and guide their actions based on your preferences. For example, they can help the AI write emails in your style or take on roles like a software engineer or copywriter.
Kerlig supports over 350 AI models, including providers like OpenAI, Google Gemini, Anthropic Claude, Perplexity, AWS Bedrock, OpenRouter, and more. It also supports running local models via Ollama and LM Studio integrations.
Learn more
