SuperAPI AI Directory
Partner with us中文AI newsFavorites

All AI tools

35,738 results · Page 790 of 993

Octopal

AI agents

Octopal is a local autonomous AI agent platform whose core agent is named Octo, and it can run on a personal computer or server, interacting via Telegram, WhatsApp, or a dedicated desktop application. It uses a coordinator/worker architecture: Octo acts as a long-running coordinator responsible for conversation, memory, strategy, and routing; workers are short-lived, one-off sub-agents that execute specific tasks in isolated Docker environments, and the instance is deleted once the task is complete. This design is intended to prevent context contamination, prompt injection, and malicious environment manipulation, making powerful automation capabilities safer and more predictable. Octopal supports executing arbitrary tasks, scheduled activities, automated pipelines, web research, and can integrate with APIs, MCP, A2A, skills, and more. It supports multiple model providers, including OpenAI, Anthropic, Google Gemini, Mistral, OpenRouter, Together AI, Groq, Codex, Z.ai, Minimax, Ollama, and custom OpenAI-compatible interfaces. The project uses the MIT license, and the code is hosted on GitHub.

Read more

Octri

AI coding

Octri is a platform that generates API documentation, SDKs, MCP servers, and monitoring based on the OpenAPI specification. Users upload a specification once (supporting formats such as OpenAPI, Swagger, Postman, and AsyncAPI, automatically normalized to OpenAPI 3.1), and can simultaneously obtain four products: a documentation site (AI-generated first draft, editable, custom domain), client SDKs in 10 languages (configurable to team style and automatically published to various registries), an MCP server for AI agents (one-click installation with npx @octri/mcp, converting API endpoints into callable tools), and production monitoring (tracing SDK or backend errors to specific commits, with distributed tracing, alerts, and GDPR compliance). The platform also provides specification readiness scoring and tools for migrating from existing configurations. In terms of pricing, documentation, SDKs, and monitoring are included in one plan, with additional SDK languages as a flat $50 per month add-on; complete details on specific plan tiers and prices are not yet available.

Read more

Odysseus AI

AI agents

Odysseus AI is a privacy-focused AI workspace for deep research and autonomous agent workflows. Users can place materials such as PDF, Markdown, and TXT files, URLs, and notes into an isolated workspace, where research agents organize them into reports with source-traceable citation tables. The product routes models through OpenRouter, supports bringing your own OpenRouter API Key (BYOK), and claims to use zero data retention (ZDR) routing and not to train models on user data. Agents have no access by default and can only read files, URLs, and tools explicitly authorized for the current task, and support extending context through custom MCP tools. The workspace also provides isolated memory that stores sources, notes, and past reports for reuse in subsequent tasks. Official typical scenarios include academic and technical literature reviews, competitor and product benchmarking, market intelligence and OSINT, and due diligence and investment research. The platform offers tiered subscription plans, including quotas for research report counts, agent run counts, saved sources and notes, and workspace storage space, with annual payment by default, switchable to monthly payment, and claims to provide a 30-day money-back guarantee. Information such as whether it is self-hosted and specific price figures is not fully provided officially and must be based on the official page and repository.

Read more

Oeneo AI

AI design

Oeneo AI is an AI image and video creation workbench that runs in the browser, integrating image generation, video generation, and post-editing tools into a single interface, all sharing one credits balance. Users can generate images from text or reference materials, then turn images into videos, and complete post-processing such as background removal, upscaling, denoising, vectorization, as well as video upscaling, extension, keying, and motion transfer within the same workspace. The platform brings together multiple models including GPT Image, Nano Banana, FLUX, Kling, Seedance, MiniMax, Vidu, and others, allowing users to view the credits cost before submitting, with failed tasks automatically refunding credits. Free registration grants 10 starting credits valid for 365 days; paid plans provide more credits as well as private generation options.

Read more