SuperAPI AI Directory
Partner with us中文AI newsFavorites

All AI tools

35,721 results · Page 140 of 993

minimind

AI developer platforms

MiniMind is an open-source project for training an ultra-small language model from scratch. It aims to lower the barriers to learning and reproducing large language models (LLMs), enabling even ordinary personal GPUs to complete training quickly. The smallest version of the project's mainline has about 64M parameters, roughly 1/2700 the size of GPT-3, and can be reproduced on a single 3090 with only about 3 yuan in cost and 2 hours of training time. The project open-sources a minimalist model architecture and a complete training pipeline, covering the full process code for MoE, data cleaning, pretraining, supervised fine-tuning, LoRA, RLHF (DPO), RLAIF (PPO/GRPO/CISPO), Tool Use, Agentic RL, adaptive thinking, and model distillation. All core algorithms are implemented natively in PyTorch and do not rely on third-party high-level abstraction interfaces. This is not only an open-source full-stage reproduction project for large language models, but also a tutorial for LLM beginners and practitioners. It is open-sourced under the Apache 2.0 license and is completely free.

Read more

MisakaNet

AI developer platforms

MisakaNet is a "failure experience library" for AI coding agents, using a Git repository as the source of truth and Markdown files to record debugging failures and fix experiences. When an agent encounters an error, it can first search the included experience lessons to avoid repeating pitfalls; when no match is found, it can submit a new failure through the intake interface for later inclusion. It provides agent-native interfaces such as an MCP server (7 tools), WebMCP, llms.txt, and A2A discovery. Retrieval uses BM25 implemented with the Python standard library, with no vector database, no embedding model, zero third-party dependencies by default, and can run offline. The project also provides a GitHub Action, a DeepSeek Harness plugin, and install/verify scripts, under the Apache-2.0 license.

Read more

Mito Health

AI healthcare

Mito Health is a platform that provides personalized health test recommendations and membership-based services. After users fill out a 60-second questionnaire, they can receive testing suggestions targeted at their personal issues: which tests to take, which can be skipped, and an overview of costs. The platform aggregates more than 1,000 biomarkers and test panels, offers testing at the "cost price" it labels, and helps compare prices and book nearby laboratories or at-home blood draws. Test results are reviewed by clinicians, and a personalized action plan covering supplements, exercise, nutrition, and sleep is provided. The membership fee is $9/month, unlocking member pricing, trend tracking, an AI health coach, year-round doctor support, and a personal health record library. The platform also sells clinician-reviewed supplements and provides consultations. The platform does not replace a doctor's diagnosis, and tests must be reviewed and ordered by partner doctors.

Read more

mkp

AI developer platforms

MKP is a Model Context Protocol (MCP) server for Kubernetes, developed by StacklokLabs. It enables LLM-driven applications to interact directly with Kubernetes clusters, providing tools via the MCP protocol such as listing and getting resources, creating and updating, deleting, and executing commands in Pods, without relying on external CLIs such as kubectl and helm. MKP is implemented in Go, calls the Kubernetes API directly, supports any resource type and CRDs, has built-in rate limiting, defaults to read-only, and can enable write operations via --read-write.

Read more

MLflow

AI developer platforms

MLflow is the largest open-source AI engineering platform, designed for agents, LLMs, and machine learning models. It helps teams debug, evaluate, monitor, and optimize AI applications: it captures complete traces of LLM applications and agents, built on OpenTelemetry and supporting any LLM provider and agent framework; it runs systematic evaluations, tracks quality metrics, and catches regressions; it provides 50+ built-in metrics and LLM judges, and also supports custom ones. It also includes prompt version management, testing, deployment, and automatic optimization, a unified LLM provider API gateway, and the MLflow Agent Server for one-click agent deployment. MLflow is licensed under Apache 2.0, 100% open source, permanently free, with no vendor lock-in, and can be used with any cloud, framework, or tool. It has been validated at scale in Fortune 500 companies and thousands of teams, with 20K+ GitHub stars, 900+ contributors, and over 30 million monthly downloads. It has built-in support for 100+ AI frameworks including LangChain, OpenAI, PyTorch, and more, and you can start a server and quickly get started with tracing with a single command.

Read more

mongodb-lens

AI developer platforms

MongoDB Lens is a locally running Model Context Protocol (MCP) server that lets users access MongoDB databases through natural language with a large language model (LLM), performing operations such as queries, aggregations, and performance optimization. It provides a complete MongoDB management toolset, covering operations for databases, collections, indexes, users, transactions, full-text search, geospatial queries, GridFS, and more, with a built-in confirmation mechanism to protect destructive operations. It supports installation via NPX, Docker Hub, source code (Node.js/Docker), and other methods, can be customized through configuration files, environment variables, and multiple connection aliases, and is compatible with modern and older versions of MongoDB.

Read more